I am very glad that someone still remembers electronic tubes. We are about the same age and the nostalgia for electronic tubes has not passed. Congratulations for the elegant and concise way in which you explained the replacement of the electronic tube.
This right here is the "hook". Many are qualified to repair, not so many enjoy the experiment and reverse engineering that is essential to UNDERSTAND why the circuit works. I can totally relate to the 100 in 1 kit reference as that was instrumental to my career. Bravo Manuel. Keep up the great work and joy of discovery!
Fascinating! I love this experiment. Absolutely excellent! Springs and wires indeed - I still have my Science Fair 150-in-1 electronics project kit that I got for Christmas as a kid in the early 80's!!! As an aside, I found on eBay a set of five solid state converted tubes for the older Trans-Oceanic Zenith radios. They were not cheap and coming from the USA made them quite expensive but some days later I decided I might give them a go but they had sold and none have come up since. My jaw dropped when I saw them appear on a Shango video three or four weeks later!! He had beat me to it and got that very set I had on watch!!
I got my electronics set at about age 12, I think. Absolutely fascinating! No idea where it is now 😊 Which video of Shangoo? Do you know? Would like to see them in action for myself.
@@electronicsoldandnew It's a long video!!! What's fascinating is the person who made them got the PCB boards back into a glass tube! And there are some issues, which are interesting to see: ruclips.net/video/3QvYSFd5qG0/видео.html
Manuel, I loved this. Granted, THESE tubes are still plentiful, but some are not, and this is a "Proof of concept" to replace the old tubes. You could even 3D print an envelope that looks like a tube, and install an LED instead of the filament, so it glows. WHAT A CONCEPT!! BRAVO!!
Also the low hanging fruit (rectifier tubes) has already been picked, so someone has to take on analysis of tubes that are actually in the signal path.
Very nice work !!!!!! I enjoyed the explanation of how the tube worked, and what you had to do to make it work with solid state. Again, it was fun to watch. Keep up the good work !!!!!!!!
Awesome video and well done with the explanations. Been tinkering with electronics before I was 8 years old I'm now 54 and loved the video thank you for putting in the hard work. 73 N8MXA Ohio USA
I enjoyed watching and listening and thank you for inviting us into your experiment. There is one remaining question and that is the cathode resistor, You need about 0,4 milliamp. And to obtain that, you need about a 10k resistor. So far so good. You bring that resistor into the socket. But you, or someone else, next time perhaps will need 1 milliamp or 2 milliamp and thus will need another resistor value. In that case this particular solid state socket won’t work properly. Another resistor value will be needed. So the solid state replacement isn’t interchangeable flawlessly. It will need a dedicated approach for every application. And as you introduce a carhode resistor that was zero in the original situation, the AC amplification will drop. I didn’t hear you talking about that.
Welcome aboard 😊 I did mention that the resistor value was found by experiment, and would probably differ from care to case. This wasn’t meant as a general replacement, but rather as a personal attempt to get it to work. Also, in a situation in which the triode is cathode biases, a resistor will already exist, so I’m not sure how this thing would bias.
@@electronicsoldandnew Yes indeed you mentioned it. but it raised a question. Would it be possible to construct a real solid state pin to pin replacement of a valve (or tube) that would be a universal and real equivalent? And secondly. I find a lay out without a cathode resistor as surprising as you. You didn’t use the simplest scheme to test your knowledge. But it was very entertaining to see.
I think that a direct replacement for a triode that uses fixed or cathode bias is possible, but not guaranteed because these fets are notoriously different, even from the same batch. Gris leak bias is a different story 😊
Glad people are finding more tubes we can replace with diodes and lnd150. One day our tube selection will be depleted 😢 Wish I would have got in to the hobby ten years before, so I could have started on my tube collection earlier. So many tubes are so expensive today!
Absolutely brilliant video! I'm an amateur radio operator and it's great to witness (on the scope) the peaking of the IF and the resulting clean output. Also a great lesson about reading schematics and comparing how tubes and FETs work.
"It's all about the tubes." Quite so. But even so, I loved watching this unfold. I liked your explanations regarding the circuit biasing and the FET. Hands on learning at its best. First rate video.
Very interesting little experiment. After retirement I dove into tube amps and built myself a clone Deluxe Reverb (because I always wanted one) which lead me to solid state receivers, buying them cheap and broken and bringing them back from the dead. I had no idea at the time what a rabbit hole I was falling through but it has been educational, satisfying and, most importantly, keeps my mind active. FET's are still a bit of a mystery to me, but I'll get there in time. Thanks for sharing!
It's good hearing you start out with a warning about the tube B+ voltage. Many years ago, while in high school, I was at a bench doing lab instrument repair while another student sat next to me working on a tube based CB transceiver. The chassis was frightening as it looked as if someone dropped components and soldered them where they landed. After a time, this student suddenly gripped the underside of the chassis with both hands. His body went limp and he slid off the stool falling to the floor, both hands still gripping the chassis, the bench mains power remained hot. Right behind me and over my right shoulder was the power kill switch for the lab. Just moments from when the guy landed on the floor, I hit the kill switch. The attending teacher and other students in the class had no idea what transpired. Fortunately, the student on the floor recovered immediately upon power having been shutdown, but both his hands were covered with burn blisters and skin punctures from the intensity of gripping the chassis.
Yes, one can quite easily forget the seriousness of a shock. I’ve had a few, and have been fortunate, but it is always important to remember and remind others of the dangers.
I was amazed at this retrofit, and got almost as much joy watching and learning, as you did doing the actual experiments. this would come in handy in future, when these Vacuum tubes are no longer available. if all the different types could be refitted with modern components, the lack of tubes would no longer be an issue. 😀 I gave you a sub and a huge like! 🤗
One thing you didn't address was power dissipation. Putting 100 volts across the FET puts a noticeable power dissipation. You can use your infrared thermometer to see the heat generated and see if it is exceeding the rating of the FET. While the FET will work with major overpower dissipation, the lifespan will be shortened. Sometimes to just a few minutes.
I did check. The datasheets show that it is well within the “safety margin”, but for fets that could be misleading. However, it’s passing less than 0.3mA, so the power dissipation is about 0.3W. The specs say 0.7W (on a cloudy day 😊)
@@electronicsoldandnew .7 watt AT 25C CASE TEMPERATURE. This means you must use refrigeration to maintain that 25C. Typical trick spec in almost all products!
True. That’s why I mentioned a cloudy day 😊 Anyway, I left it active for a couple of hours and the temp didn’t go high at all. The current must be very small indeed. The spec sheet is quite good on this component: ww1.microchip.com/downloads/en/DeviceDoc/LND150%20C041114.pdf
Nicely done - the simplicity is what makes it great. I looked at the spec sheet - looks like the FET may just be fast enough to also replace the EF89 IF amp, but would have to work at less plate current that the original tube. Would be interesting to try.
@@electronicsoldandnew Yes! That high voltage low current fet you used sure has possibilities - really similar to a tube other than lack of grid current.
You could use an RF JFET then cascode it with the LND150, would stand a better chance at 10.7 MHz. Set the current so that nothing blows up. Any takers?
@@stevehead365 Thats an intriguing idea. In this radio looks like the cathode is grounded making things easy. Maybe use a voltage tripler to get around -20V for the FET half of the circuit. Would be nice to make it more universal and not depend on the cathode being grounded. Something to think about. Would be great fun to play around with.
Hi Manuell Thanks for your many video's which I enjoy watching. This is just the information I have been loking for since I have some 1910-20' radio with bad tubes that I cannot replace easely so its super if I can make some tube replacement on a socket and just swap it out if I come across some working tubes. Have a nice day
I think the FET will not work very well at higher frequencies, but I was not able to see a frequency limit value on the datasheet. It did say "rise time" of .45 microseconds .... which suggests an upper frequency limit of 1-2Mhz ... not quite good enough for an AM radio? Perhaps someone can explain.
@@electronicsoldandnew I think the FET will not work very well at higher frequencies, but I was not able to see a frequency limit value on the datasheet. It did say "rise time" of .45 microseconds .... which suggests an upper frequency limit of 1-2Mhz ... not quite good enough for an AM radio? Perhaps someone can explain.
Nice and simple when you have the info available to correctly bias the DMOS FET, in fact I was surprised just how straightforward it is. Might be worth building the SM version just to use as a swap out to prove if a valve has failed. Of course it this was a radio with series strung heaters, that would have to be accounted for. I watched Shango swap the oscillator valve in the RF/IF stage of I think a radio - it may have been a TV - with a solid state valve that a guy had designed and sells. In fact I think he replaced the whole IF chain. I think it's a practical solution for those valves that just aren't available or are hideously expensive due to scarcity. But a great experiment to try even if you don't intend on replacing the valves. Great video.
I saw Shango’s video. He did in fact replace all the tubes. I didn’t think of the “check if the tube was working” aspect. May be an idea. I think the most useful replacement would be the magic eye. They are getting vvvveeerrryyy expensive. I saw someone do it before but don’t think they’re selling them. They did in fact do the best possible, so I’m not tempted to try to improve it, because I don’t think mine would be better.
Wow. I spent years trying to find the transistor. I used Shottky diodes and they work fine. I must find this transistor and try it. Wonderful job. I made experiments with russian subminiature tube as triode. It worked too.
Yesterday I received the ordered MOSFET's LND150N3-G from Farnell and instantly I made my own replacement for the EABC80. Hmmm, I had to change the sourceresistor to 1kOhms to make it work fine. In accordance to the european tube type code I call it "OABC80". The "O" as the first Letter stands for the heater and means "Cold Cathode". Many thanks for this inspiration to replace a tube by semiconductor circuitry!
Very interesting topic, when I had involvement in TV repair as a kid I remember tube replacements were available through willow vale and cpc where they used bipolar high voltage mje340 and a few bits for baising and completing the series heater chain, for pcl86 tubes
I really enjoyed your video. I am 70 years old and have often wondered if it was possible to convert a tube to a solid state alternative. Thank you for providing the proof of concept. I personally would have to take it further and mounted the components in a cut off old glass valve complete with orange LED for the "filament." Then glued the glass back together for completion. 😊 By the way would it be possible to do something similar with any other tubes? I have subscribed and look forward to your future episodes. 👍🪫
Welcome aboard 😊 I believe that all the tubes can in fact be replaced with solid statesequivalents, but I’m not sure just how well they would work. I know that others have done this, but it involves actually replacing the B+ supply to a lower one, which, in my opinion is just too much altering of the original to be feasible.
Awesome video. Please keep these types of videos coming. What better way to learn how a tube works than seeing you replicate its parts and then demonstrating that it works! I see that you couldn’t resist using your new 3D printer for your voltage multiplier. It looks great. When I need a high impedance input for DC I use my Micronta FET multimeter which has a 10 Mega Ohm input impedance but I always use the voltage multiplier when measuring AC since both analog meters (Simpson and Micronta) have low AC input impedance. Have you kept the input impedance on your multiplier to 4 Mega Ohm or have you increased it? Thanks Manuel.
No, could really not resist creating the new enclosure 😊 I left the multiplier input impedance as it was, since I have not needed a higher one to date. That could change in the future.
I truly understand your fascination - because that is exactly what it is : fascinating ! I could find many reasons for substituting "real" tubes with a solid state version. If you go back to 20'es, 30'es, 40'es, 50'es era of vacuum tubes, these are becoming more rare by the day, if replacement is needed for repair or restoration. Besides you have a wide choice of the active element, as High voltage FETs are cheap standard components these days. Thanks very much for this highly interesting video. 🤩
I have often wondered about this. It is nice to see that it actually works because someday if these tubes are ever not available anymore it's good to know that you can still get the radio to work.
Hey Manuel, I really liked that. As you said, not to replace tubes with semiconductors. I've been wondering what you showed for a long time. Thank you for this interesting video.
EXCELENTE TRABAJO!!!. Yo sustitui todas las valvulas de un Philips 2514 (1928, radio de mis tias-abuelas) por las octales/novales. Las he guardado como reliquias venerables (E415, B409 ...)
Manuel, your experimenting reminds me of some of the early magazine articles transistor theory back in the 50s. Might be interesting to look into some of those old articles. At the escalating tube prices your ckt mite be good way to keep that cheap AM radio working on a limited budget.
A 500V small FET is quite interesting. First time YT offered your channel, there are likely still other restoration channels that are hidden in the algorithms noise floor.
Putting a resistor at the source creates a feedback circuit because the audio is going to affect the source voltage, a higher current raises the voltage and makes the gate more negative with respect to the source. This is not the open loop amplification the EABC80 had. To get open loop amplification put a capacitor between source and ground. May not work better though but it is closer to how the tube worked. Just saying.
Adding a source decoupling capacitor occurred to me also, but in this case I think @stevehead365 reply sums it up. 10k may seem a high resistance for a source resistor but not compared with the drain (anode) load resistor of 220k.
Thank you for your crazy experiments. Good job! Here in Russia we can find eabc80 and its clones hardly. But germanium diodes can be obtained for a zero price.
Another substitute for the EABC80/6AK8 is the 6T8. I've used those and they work fine. I often find when checking the EABC80's, the diode sections show weak on my tester. But the tube seems to perform fine in the radio.
so good you had great results, wish youtube would let me give you 10,000 likes , next step is an output pentode challenge is on ! lol i had just missed the valve era when i was learning so i would have never understood that preamp bias system ,thats really wierd , iam so used to seeing just a cathode resistor with ac bypass cap but i can see the cathode being shared with 2 detection diodes prevents them doing it that way
I was also baffled when I saw this biasing system for the first time, but it actually makes sense. I finished varsity in 1985, so also missed the tube era, but I find it fascinating.
I had an Ekco radio I repaired recently, with a UABC80 in it (it is a no mains transformer set). I found the FM didn't work properly, so as a temporary fix I just put a couple of silicon diodes across the diodes of the FM detector part of the valve. It worked, but I got a lot lower volume. Later, I remembered that I actually had another UABC80 in my junk box, so problem solved! But yes, it was fun to experiment.
That UABC80 needs to stay in place as it’s probably a series heater string. You could have got a better result if you’d actually “cut out” the connections to the tube diodes and then places the silicon diodes in the circuit. An alignment of the final IF transformer stage would probably have brought the level back. However, that’s all academic. A replacement tube is the best solution 😊
Thank you Manuel. I notice the EABC80 is somewhat rare so a solid state equivalent could be a good idea. The gain of the triode section is around 50 to 70. The FET gain should be somewhat equivalent because the audio stage has a feedback loop around it. Also for this reason the gain of the fet should be measured by checking the gate input and the drain output The germanium diodes will give less distortion on weak am signals
That was huge fun. You'll be re-stuffing old valves next - drill a small hole in the base where it can't be seen to equalise pressure, a diamond saw to remove the base from the "bell" and then replace the innards, not with SMDs and a PCB but old school wired components tastefully arranged as an artistic showpiece. Don't forget to add an LED and dropping resistor across the 6.3VAC terminals to simulate the heater glow and illuminate the internal artwork. But you might have to drop @glasslinger a line for tips on reattaching the bell and simulating an uncontaminated getter though. 🤣 I was wondering if Schottky barrier diodes would work but maybe not as for a low forward voltage the peak reverse voltage might be an issue?
Hello i replace sometimes EABC80 with 3x1N5711 diodes and one russian 6N2P double triode. One section for preamp an one section for Phono preamp (bluetooth)works well and a radio maintains retro look.
Muito bom meu caro. Parabéns pelos excelentes trabalhos que tem partilhado, e por manter o "espírito inventivo" que tão bons momentos e satisfação proporciona (a quem pratica ... e a quem assiste ! ).
You are thee one person to Defy Murphy,s Law and be a Winner , P S We had a beautiful time recently on the Island walking Up those hilly streets was a Killer but good for an Eighty year old ,
Madeira is just lovely. We found ourselves huffing and puffing with the hills and frankly a bit lost but stumbled upon a corner cafe that served a Madeiran hot dog, at that moment it was the best thing I've ever eaten!
interesting .The first solid state vale i came across was a PL82T used in the CSA panel of the PYE hybrid colour tv when PL802 becam scarce . shangoo ha a 4 valve portble radio (useing the 1.5 volt valves) he managed to find solid state replacments for all 4 valves.
A 12AX7 replacement is not far from the replacement shown here for the triode section. IIRC one triode of the 12AX7/ECC83 and triode of the EABC80 are very similar in characteristics. So the receipe for a semiconductor 12AX7 is shown here. BUT just getting it to amplify is just 3/4 of the job. The distortion on tubes like 12AX7 near maximum peak to peak output is different from semiconductor solutions. And overdriving tubes and its semiconductor replacements gives different results. So it remains to be seen if especially for the guitar amp or audiophile community semiconductor replacements could find acceptance.
That’s why I believe that tube amp enthusiasts will undoubtedly not favour any such substitution. I have many guitarist friends who would rather play a tambourine than give up their tube amps for solid state ones 😊
Hi, Charles Smith is one of the pioneers (not the first) to practice this exercise (R392 Solid State Conversion), Theo Faber followed him in this exercise (Racal 17 transistorized). 73
I would expect that difference between germanium and silicon diode can be heard when there is a weak signal, especially on AM. FM always need some amplitude for ratio detector to start working. Diodes in ratio detector in EABC80 or EAA91 were replaced by 2 germanium ones soon during 60s.
I’ve been temped to get my hands on one of those nutubes, but they are quite expensive. They used them to emulate the 12ax7 quite well. You could create a “tube preamp” as a stompbox.
Nice job !! .....You may be losing a few audio Signal volts across the 10K source resistor Without the traditional 1uf Bypass capacitor....Simple & compact to add...
The end result is there is a viable way to replace a tube with solid state components should it become necessary. It would be interesting to see if every tube could have a replacement or whether there are some limitations. As a musician I can see value in these replacements as many tubes have become quite costly, and some outright scarce, and this could be an alternative.
Please please. Try to do a section on replacing EL84 with solid state. The tubes are expensive and I restore ideally and cannot add the cost of these tubes to my restoration. The N-mosfet should work just as well. Just the calculation in e.g. a philips radio.
Where I wouldn't mind outright replacing tubes is at the AC-to-DC stage. A lot of power could be saved there without having any meaningful effect on the circuit downstream of it, unless you are deliberately using power sag as a source of distortion like Eddie Van Halen did.
Agreed. The rectifier tube is one of the most common replacements, and power sag … you need deafening levels to achieve that. My ears wouldn’t handle it 😊
@@electronicsoldandnew Wrap it in a transparent case so it resembles a tube, and stick a little neon bulb in there to confirm B+ is good, and the only difference most people will notice is that it seems to warm up instantly.
Hi Manuel, this video made me decide to write you about my recent idea, inspired by the problem i faced with few of my stereo radios, they need magic eye emm803, which is hard to find and very expensive. I made a simple circuit which i tested and working fine consisted of em84(6E2) green led and few components. I wanted to ask you if you agree to develop a pcb that can be fited into this chinese socket, so em84 goes into the socket and socket goes to the original socket of the radio without messing any components in the radio itself. I will email you a picture of the schematic and the final tube i made... so i look forward to hear what you think. Greetings from Bulgaria.
That sound interesting. I’d love to give it a try. Email me the details. By the way, if I do the video, others will be able to order their own PCBs with no benefit to you or me, so I hope you don’t mind. Contrary to popular belief, I don’t get paid commissions on PCBs ordered from my sponsor. I do get my own for free and they sponsor the videos, but that’s about it. I don’t want to make a business of this. It’s too much fun as a hobby 😊
a question ive allwas wondered with detection diodes is why dont designers give them a little bit of pre bias to help them out ,something low enough to not have any ambient temperature effects something in the ballpark of 200mv for silicon 100mv for germainium so still cut of by a big margin but a bit more ready for signal, i want to find out for myself right now but my "man lab" is currently repurposed as a bedroom grrr!
That is a good way to make them instantly sensitive, as is done with transistors. Should be easy in theory, but not so simple in practice se there are oppositely biased diodes, so opposing supplies would need to be designed, and could make the circuit a little messy.
Maybe it is time for a 12AX7 semiconductor replacement in both audio sweep/pink noise spectrum analysis and blindfold listening test in a guitar amp or stereo console, with respect to the tube.
I am very glad that someone still remembers electronic tubes. We are about the same age and the nostalgia for electronic tubes has not passed. Congratulations for the elegant and concise way in which you explained the replacement of the electronic tube.
👍 it’s a pleasure
This right here is the "hook". Many are qualified to repair, not so many enjoy the experiment and reverse engineering that is essential to UNDERSTAND why the circuit works. I can totally relate to the 100 in 1 kit reference as that was instrumental to my career. Bravo Manuel. Keep up the great work and joy of discovery!
Thank you, and same to you 👍
Fascinating! I love this experiment. Absolutely excellent! Springs and wires indeed - I still have my Science Fair 150-in-1 electronics project kit that I got for Christmas as a kid in the early 80's!!! As an aside, I found on eBay a set of five solid state converted tubes for the older Trans-Oceanic Zenith radios. They were not cheap and coming from the USA made them quite expensive but some days later I decided I might give them a go but they had sold and none have come up since. My jaw dropped when I saw them appear on a Shango video three or four weeks later!! He had beat me to it and got that very set I had on watch!!
I got my electronics set at about age 12, I think. Absolutely fascinating! No idea where it is now 😊
Which video of Shangoo? Do you know? Would like to see them in action for myself.
@@electronicsoldandnew It's a long video!!! What's fascinating is the person who made them got the PCB boards back into a glass tube! And there are some issues, which are interesting to see: ruclips.net/video/3QvYSFd5qG0/видео.html
The guy is called Bill Turner.
@@electronicsoldandnew Well that's now got me on Google digging around!!! 😂
Manuel, I loved this. Granted, THESE tubes are still plentiful, but some are not, and this is a "Proof of concept" to replace the old tubes. You could even 3D print an envelope that looks like a tube, and install an LED instead of the filament, so it glows. WHAT A CONCEPT!! BRAVO!!
😊 thanks
Could use a miniature automotive lamp instead of an LED to enable series string usage.
Also the low hanging fruit (rectifier tubes) has already been picked, so someone has to take on analysis of tubes that are actually in the signal path.
👍
That's why I enjoy your channel, you're often exploring something different, and I am not an engineer so it's a welcome nudge outside my comfort zone.
👍
Very nice work !!!!!!
I enjoyed the explanation of how
the tube worked, and what you had
to do to make it work with solid state.
Again, it was fun to watch.
Keep up the good work !!!!!!!!
👍
I love the "because we can" video's. Keep doing what you love
👍
Damn, that is a Portuguese station...!!!! 🤣🤣... eu devia ter calculado ´´M. Caldeira´´, excelente trabalho ...!!!
Obrigado 😊
Awesome video and well done with the explanations. Been tinkering with electronics before I was 8 years old I'm now 54 and loved the video thank you for putting in the hard work. 73 N8MXA Ohio USA
My pleasure
I enjoyed watching and listening and thank you for inviting us into your experiment. There is one remaining question and that is the cathode resistor, You need about 0,4 milliamp. And to obtain that, you need about a 10k resistor. So far so good. You bring that resistor into the socket. But you, or someone else, next time perhaps will need 1 milliamp or 2 milliamp and thus will need another resistor value. In that case this particular solid state socket won’t work properly. Another resistor value will be needed. So the solid state replacement isn’t interchangeable flawlessly. It will need a dedicated approach for every application. And as you introduce a carhode resistor that was zero in the original situation, the AC amplification will drop. I didn’t hear you talking about that.
Welcome aboard 😊 I did mention that the resistor value was found by experiment, and would probably differ from care to case. This wasn’t meant as a general replacement, but rather as a personal attempt to get it to work. Also, in a situation in which the triode is cathode biases, a resistor will already exist, so I’m not sure how this thing would bias.
@@electronicsoldandnew Yes indeed you mentioned it. but it raised a question. Would it be possible to construct a real solid state pin to pin replacement of a valve (or tube) that would be a universal and real equivalent? And secondly. I find a lay out without a cathode resistor as surprising as you. You didn’t use the simplest scheme to test your knowledge. But it was very entertaining to see.
I think that a direct replacement for a triode that uses fixed or cathode bias is possible, but not guaranteed because these fets are notoriously different, even from the same batch. Gris leak bias is a different story 😊
Glad people are finding more tubes we can replace with diodes and lnd150. One day our tube selection will be depleted 😢
Wish I would have got in to the hobby ten years before, so I could have started on my tube collection earlier. So many tubes are so expensive today!
True.
I love this stuff too! Especially with modern test gear you can get a very granular view of exactly how this stuff works. So cool!
👍
Absolutely brilliant video! I'm an amateur radio operator and it's great to witness (on the scope) the peaking of the IF and the resulting clean output. Also a great lesson about reading schematics and comparing how tubes and FETs work.
Thank you 👍
"It's all about the tubes." Quite so. But even so, I loved watching this unfold. I liked your explanations regarding the circuit biasing and the FET. Hands on learning at its best. First rate video.
Thank you
Very interesting little experiment. After retirement I dove into tube amps and built myself a clone Deluxe Reverb (because I always wanted one) which lead me to solid state receivers, buying them cheap and broken and bringing them back from the dead. I had no idea at the time what a rabbit hole I was falling through but it has been educational, satisfying and, most importantly, keeps my mind active. FET's are still a bit of a mystery to me, but I'll get there in time. Thanks for sharing!
That rabbit hole has caught many of us 😊
A bit of fun there Manuel. I was surprised the silicone diodes worked so well 🙂
So was I 😊
Wonderful experiment/demo . Most interesting. Thank you .
My pleasure
This is so incredibly cool! It's one thing to know that it's possible in theory, and quite another to see it work. Brilliant job.
Now do an EL84.
Thank you 😊
Wonderful experiment. I really enjoy this video. Very useful 😊 thanks !!
My pleasure
Excellent, Manuel. Do more off this kind of experiments.
👍
It's good hearing you start out with a warning about the tube B+ voltage.
Many years ago, while in high school, I was at a bench doing lab instrument repair while another student sat next to me working on a tube based CB transceiver. The chassis was frightening as it looked as if someone dropped components and soldered them where they landed. After a time, this student suddenly gripped the underside of the chassis with both hands. His body went limp and he slid off the stool falling to the floor, both hands still gripping the chassis, the bench mains power remained hot. Right behind me and over my right shoulder was the power kill switch for the lab. Just moments from when the guy landed on the floor, I hit the kill switch. The attending teacher and other students in the class had no idea what transpired. Fortunately, the student on the floor recovered immediately upon power having been shutdown, but both his hands were covered with burn blisters and skin punctures from the intensity of gripping the chassis.
Yes, one can quite easily forget the seriousness of a shock. I’ve had a few, and have been fortunate, but it is always important to remember and remind others of the dangers.
Very entertaining and very informative. Thanks for sharing
My pleasure
I can fully understand the joy you are having! Very nice video! 😀
Thanks
...and a beautiful time was had by all. Thanks, Manuel. 🙂
My pleasure 😊
I was amazed at this retrofit, and got almost as much joy watching and learning, as you
did doing the actual experiments. this would come in handy in future, when these Vacuum
tubes are no longer available. if all the different types could be refitted with modern components,
the lack of tubes would no longer be an issue. 😀 I gave you a sub and a huge like! 🤗
Thank you
One thing you didn't address was power dissipation. Putting 100 volts across the FET puts a noticeable power dissipation. You can use your infrared thermometer to see the heat generated and see if it is exceeding the rating of the FET. While the FET will work with major overpower dissipation, the lifespan will be shortened. Sometimes to just a few minutes.
I did check. The datasheets show that it is well within the “safety margin”, but for fets that could be misleading. However, it’s passing less than 0.3mA, so the power dissipation is about 0.3W. The specs say 0.7W (on a cloudy day 😊)
@@electronicsoldandnew .7 watt AT 25C CASE TEMPERATURE. This means you must use refrigeration to maintain that 25C. Typical trick spec in almost all products!
True. That’s why I mentioned a cloudy day 😊 Anyway, I left it active for a couple of hours and the temp didn’t go high at all. The current must be very small indeed. The spec sheet is quite good on this component: ww1.microchip.com/downloads/en/DeviceDoc/LND150%20C041114.pdf
@@glasslinger one these days I to talk to you.
@@electronicsoldandnewyou're going to get somebody killed you with that would b +voltage
Nicely done - the simplicity is what makes it great. I looked at the spec sheet - looks like the FET may just be fast enough to also replace the EF89 IF amp, but would have to work at less plate current that the original tube. Would be interesting to try.
I think one can set this fet to replicate quite a few triodes and even lentores in triode mode. Nothing that a little experimentation won’t solve 😊
@@electronicsoldandnew Yes! That high voltage low current fet you used sure has possibilities - really similar to a tube other than lack of grid current.
👍
You could use an RF JFET then cascode it with the LND150, would stand a better chance at 10.7 MHz. Set the current so that nothing blows up. Any takers?
@@stevehead365 Thats an intriguing idea. In this radio looks like the cathode is grounded making things easy. Maybe use a voltage tripler to get around -20V for the FET half of the circuit. Would be nice to make it more universal and not depend on the cathode being grounded. Something to think about. Would be great fun to play around with.
Hi Manuell
Thanks for your many video's which I enjoy watching.
This is just the information I have been loking for since I have some 1910-20' radio with bad tubes that I cannot replace easely so its super if I can make some tube replacement on a socket and just swap it out if I come across some working tubes.
Have a nice day
👍 good luck with your efforts
I think the FET will not work very well at higher frequencies, but I was not able to see a frequency limit value on the datasheet. It did say "rise time" of .45 microseconds .... which suggests an upper frequency limit of 1-2Mhz ... not quite good enough for an AM radio? Perhaps someone can explain.
@@electronicsoldandnew I think the FET will not work very well at higher frequencies, but I was not able to see a frequency limit value on the datasheet. It did say "rise time" of .45 microseconds .... which suggests an upper frequency limit of 1-2Mhz ... not quite good enough for an AM radio? Perhaps someone can explain.
Also look forward to learning from anyone out there who has an answer
Nice and simple when you have the info available to correctly bias the DMOS FET, in fact I was surprised just how straightforward it is. Might be worth building the SM version just to use as a swap out to prove if a valve has failed. Of course it this was a radio with series strung heaters, that would have to be accounted for. I watched Shango swap the oscillator valve in the RF/IF stage of I think a radio - it may have been a TV - with a solid state valve that a guy had designed and sells. In fact I think he replaced the whole IF chain. I think it's a practical solution for those valves that just aren't available or are hideously expensive due to scarcity. But a great experiment to try even if you don't intend on replacing the valves. Great video.
I saw Shango’s video. He did in fact replace all the tubes. I didn’t think of the “check if the tube was working” aspect. May be an idea.
I think the most useful replacement would be the magic eye. They are getting vvvveeerrryyy expensive. I saw someone do it before but don’t think they’re selling them. They did in fact do the best possible, so I’m not tempted to try to improve it, because I don’t think mine would be better.
@@electronicsoldandnew I'll look out for the magic eye video.
Wow. I spent years trying to find the transistor. I used Shottky diodes and they work fine. I must find this transistor and try it. Wonderful job. I made experiments with russian subminiature tube as triode. It worked too.
👍
Amazing, tube replacements using solid state components. Never thought about that.
😊
Very detailed analysis. I have always wondered about doing this to a pentode circuit
👍
What a great video ! You are making a fun hobby even better with your insight .
Had fun making it.
Awesome 👌.. The thrill
😊
Yesterday I received the ordered MOSFET's LND150N3-G from Farnell and instantly I made my own replacement for the EABC80. Hmmm, I had to change the sourceresistor to 1kOhms to make it work fine. In accordance to the european tube type code I call it "OABC80". The "O" as the first Letter stands for the heater and means "Cold Cathode". Many thanks for this inspiration to replace a tube by semiconductor circuitry!
👍 glad to have inspired you 👍
Very interesting topic, when I had involvement in TV repair as a kid I remember tube replacements were available through willow vale and cpc where they used bipolar high voltage mje340 and a few bits for baising and completing the series heater chain, for pcl86 tubes
👍
I really enjoyed your video. I am 70 years old and have often wondered if it was possible to convert a tube to a solid state alternative. Thank you for providing the proof of concept. I personally would have to take it further and mounted the components in a cut off old glass valve complete with orange LED for the "filament." Then glued the glass back together for completion. 😊
By the way would it be possible to do something similar with any other tubes? I have subscribed and look forward to your future episodes. 👍🪫
Welcome aboard 😊
I believe that all the tubes can in fact be replaced with solid statesequivalents, but I’m not sure just how well they would work. I know that others have done this, but it involves actually replacing the B+ supply to a lower one, which, in my opinion is just too much altering of the original to be feasible.
Enjoyed your inspiration and experience .........thanks
My pleasure
I learned a lot about the EABC80 while watching this. I also did not know that one can get FETs for such high voltages...
👍
I've been using Schottky diodes for the detection job, at least they are fast enough and have a low threshold voltage.
👍
Excellent detective work! If it can be done, it's you! 👍 Well done, my friend! Would like to have seen the final schematic of the solid state circuit.
👍
Awesome video. Please keep these types of videos coming. What better way to learn how a tube works than seeing you replicate its parts and then demonstrating that it works! I see that you couldn’t resist using your new 3D printer for your voltage multiplier. It looks great. When I need a high impedance input for DC I use my Micronta FET multimeter which has a 10 Mega Ohm input impedance but I always use the voltage multiplier when measuring AC since both analog meters (Simpson and Micronta) have low AC input impedance. Have you kept the input impedance on your multiplier to 4 Mega Ohm or have you increased it? Thanks Manuel.
No, could really not resist creating the new enclosure 😊
I left the multiplier input impedance as it was, since I have not needed a higher one to date. That could change in the future.
I truly understand your fascination - because that is exactly what it is : fascinating !
I could find many reasons for substituting "real" tubes with a solid state version.
If you go back to 20'es, 30'es, 40'es, 50'es era of vacuum tubes, these are becoming more rare by the day, if replacement is needed for repair or restoration.
Besides you have a wide choice of the active element, as High voltage FETs are cheap standard components these days.
Thanks very much for this highly interesting video. 🤩
Im sure someone can work up a replacement for the really old tubes. All it takes is some patience and a few gets to blow up 😊
@@electronicsoldandnew Exactly. Any experienced Radio Valve "head" can find the original valve specifications and make up a solid state replacement.
👍
I have often wondered about this. It is nice to see that it actually works because someday if these tubes are ever not available anymore it's good to know that you can still get the radio to work.
👍
Hey Manuel, I really liked that. As you said, not to replace tubes with semiconductors. I've been wondering what you showed for a long time.
Thank you for this interesting video.
👍 my pleasure
Fantastic video, two thumbs up.
Thank you
EXCELENTE TRABAJO!!!. Yo sustitui todas las valvulas de un Philips 2514 (1928, radio de mis tias-abuelas) por las octales/novales. Las he guardado como reliquias venerables (E415, B409 ...)
👍
Replicating this for the short lived ECC85 would be cool-thanks Manuel!
👍
Excellent, amazing.
I try to do something like that.
Thanks for share
👍
Manuel, your experimenting reminds me of some of the early magazine articles transistor theory back in the 50s. Might be interesting to look into some of those old articles. At the escalating tube prices your ckt mite be good way to keep that cheap AM radio working on a limited budget.
👍
Very well done. A solid state version of the ECLL800 would be a nice experiment...
Now that would be a nice challenge :)
Thanks 👍👍 i am from Morocco i love your vedio
👍
A triple-diode-triode on a tube saver base?!! Oh wow!
I bet you're one of just a few today, who've built that vacuum tube from scratch. I say!
Kudos!
Thank you 😊
Thank you , very nice. Love it.
👍
Nice project ! Clever.
👍
A 500V small FET is quite interesting. First time YT offered your channel, there are likely still other restoration channels that are hidden in the algorithms noise floor.
😊 probably a lot of channels waiting to be found. This one has been around for years.
Putting a resistor at the source creates a feedback circuit because the audio is going to affect the source voltage, a higher current raises the voltage and makes the gate more negative with respect to the source. This is not the open loop amplification the EABC80 had. To get open loop amplification put a capacitor between source and ground. May not work better though but it is closer to how the tube worked. Just saying.
👍
FETs are far from linear, so some NFB is useful as long as the voltage gain is adequate. Seems to work well here.
👍
Adding a source decoupling capacitor occurred to me also, but in this case I think @stevehead365 reply sums it up. 10k may seem a high resistance for a source resistor but not compared with the drain (anode) load resistor of 220k.
True. It’s the ratio that seems to be more crucial. With fets you never really know. It can vary greatly from one to another, even from the same pot.
Bravo, sir. Well done.
Thank you
Thank you for your crazy experiments. Good job! Here in Russia we can find eabc80 and its clones hardly. But germanium diodes can be obtained for a zero price.
👍
Another substitute for the EABC80/6AK8 is the 6T8. I've used those and they work fine. I often find when checking the EABC80's, the diode sections show weak on my tester. But the tube seems to perform fine in the radio.
👍
so good you had great results, wish youtube would let me give you 10,000 likes , next step is an output pentode challenge is on ! lol
i had just missed the valve era when i was learning so i would have never understood that preamp bias system ,thats really wierd , iam so used to seeing just a cathode resistor with ac bypass cap but i can see the cathode being shared with 2 detection diodes prevents them doing it that way
I was also baffled when I saw this biasing system for the first time, but it actually makes sense. I finished varsity in 1985, so also missed the tube era, but I find it fascinating.
@@electronicsoldandnew yes to sum it up the bias system is invisable , literally!
True
How many of the other valves in the set can be done in this way? I suppose the power usage is lower.
Probably most of them, with subtle changes to the main concept.
I had an Ekco radio I repaired recently, with a UABC80 in it (it is a no mains transformer set). I found the FM didn't work properly, so as a temporary fix I just put a couple of silicon diodes across the diodes of the FM detector part of the valve. It worked, but I got a lot lower volume. Later, I remembered that I actually had another UABC80 in my junk box, so problem solved! But yes, it was fun to experiment.
That UABC80 needs to stay in place as it’s probably a series heater string. You could have got a better result if you’d actually “cut out” the connections to the tube diodes and then places the silicon diodes in the circuit. An alignment of the final IF transformer stage would probably have brought the level back. However, that’s all academic. A replacement tube is the best solution 😊
Great video . I imagine the same could be done with the 12av6 & 6av6 tubes as well
Probably.
Soo loved this. Thanks.
My pleasure
Thank you Manuel. I notice the EABC80 is somewhat rare so a solid state equivalent could be a good idea. The gain of the triode section is around 50 to 70. The FET gain should be somewhat equivalent because the audio stage has a feedback loop around it. Also for this reason the gain of the fet should be measured by checking the gate input and the drain output
The germanium diodes will give less distortion on weak am signals
👍
That was huge fun. You'll be re-stuffing old valves next - drill a small hole in the base where it can't be seen to equalise pressure, a diamond saw to remove the base from the "bell" and then replace the innards, not with SMDs and a PCB but old school wired components tastefully arranged as an artistic showpiece. Don't forget to add an LED and dropping resistor across the 6.3VAC terminals to simulate the heater glow and illuminate the internal artwork. But you might have to drop @glasslinger a line for tips on reattaching the bell and simulating an uncontaminated getter though. 🤣
I was wondering if Schottky barrier diodes would work but maybe not as for a low forward voltage the peak reverse voltage might be an issue?
😊 don’t know if I have the patience for that. The 1N5711 Schottky is supposed to be perfect for this, so I have some on order.
Is it possible, to create a solid state replacement for a ECC85 tube? It's getting expensiver to buy a NOS tube.
I think it could actually be simpler than this one
Hello i replace sometimes EABC80 with 3x1N5711 diodes and one russian 6N2P double triode. One section for preamp an one section for Phono preamp (bluetooth)works well and a radio maintains retro look.
👍
Muito bom meu caro. Parabéns pelos excelentes trabalhos que tem partilhado, e por manter o "espírito inventivo" que tão bons momentos e satisfação proporciona (a quem pratica ... e a quem assiste ! ).
Estes vídeos são um prazer produzir. Quanto mais “estranhas” as experiências, melhor 😊
You are thee one person to Defy Murphy,s Law and be a Winner , P S We had a beautiful time recently on the Island walking Up those hilly streets was a Killer but good for an Eighty year old ,
😊 the hills can be dangerous here 😊
Madeira is just lovely. We found ourselves huffing and puffing with the hills and frankly a bit lost but stumbled upon a corner cafe that served a Madeiran hot dog, at that moment it was the best thing I've ever eaten!
😊
Very cool .. Love your channel
Thank you
Very good... great...
Thanks
interesting .The first solid state vale i came across was a PL82T used in the CSA panel of the PYE hybrid colour tv when PL802 becam scarce .
shangoo ha a 4 valve portble radio (useing the 1.5 volt valves) he managed to find solid state replacments for all 4 valves.
👍
That was really cool! Now if you just moved on to each tube one by one till you had 100% solid state- That would excite everyone.
😊
I have seen 12ax7 solid state replacements but I wonder about power sections like 6l6 and el84's and such if it is doable in a solid state format.
I’ve seen it done, but not sure how at this stage.
A 12AX7 replacement is not far from the replacement shown here for the triode section. IIRC one triode of the 12AX7/ECC83 and triode of the EABC80 are very similar in characteristics. So the receipe for a semiconductor 12AX7 is shown here.
BUT just getting it to amplify is just 3/4 of the job. The distortion on tubes like 12AX7 near maximum peak to peak output is different from semiconductor solutions. And overdriving tubes and its semiconductor replacements gives different results. So it remains to be seen if especially for the guitar amp or audiophile community semiconductor replacements could find acceptance.
That’s why I believe that tube amp enthusiasts will undoubtedly not favour any such substitution. I have many guitarist friends who would rather play a tambourine than give up their tube amps for solid state ones 😊
Very interesting topic .
👍
Hi, Charles Smith is one of the pioneers (not the first) to practice this exercise (R392 Solid State Conversion), Theo Faber followed him in this exercise (Racal 17 transistorized). 73
Just went to see that video. Really interesting 👍 thanks for that.
@@electronicsoldandnew these guys are goldmines of practice. You're welcome.
👍
Using a capacitor to AC-bypass the source bias resistor might be a good idea too.
👍
I would expect that difference between germanium and silicon diode can be heard when there is a weak signal, especially on AM. FM always need some amplitude for ratio detector to start working. Diodes in ratio detector in EABC80 or EAA91 were replaced by 2 germanium ones soon during 60s.
👍
Yes, to test the level between diodes you need to disable the AVC and use a generator with a fix level attached to the radio.
👍
Hi Manuel there are other tubes getting rare like the 1L6, experimenting with replacing these types of tubes would be more useful. Vince
Probably, but I’ve never come across one of them (I think), so I’ll leave that for someone else to delve into 😊
I believe Roland made something like this for its Blues Cube amplifier. A few years ago Korg company made a new solid state tube called NuTube
I’ve been temped to get my hands on one of those nutubes, but they are quite expensive. They used them to emulate the 12ax7 quite well. You could create a “tube preamp” as a stompbox.
cool stuff stil old and cool
👍
But can you do the triode hexode or heptode? That one seems like quite a challenge. Again, those tubes aren't used much so they are cheap but still...
Can you try using Schottky diodes for the detector experiments?
You can, yes
Nice job !! .....You may be losing a few audio Signal volts across the 10K source resistor Without the traditional 1uf Bypass capacitor....Simple & compact to add...
👍
Very cool. Now let's see if you can make a solid state 12AX7/ECC83 work in a guitar amp 😉
It’s been done 😊 ruclips.net/video/5-pLkgAKv1w/видео.htmlsi=u8GiWSxZ9V-bEw1f
@@electronicsoldandnew ahhhh, but one that sounds good and reacts like a tube with the same curves.
😊
I would have tried bypassing the 10k source resistor with a electrolytic capacitor (value decided with testing) to increase the AC gain.
The gain is very high as it is, but yes. That would be the effect.
The end result is there is a viable way to replace a tube with solid state components should it become necessary. It would be interesting to see if every tube could have a replacement or whether there are some limitations. As a musician I can see value in these replacements as many tubes have become quite costly, and some outright scarce, and this could be an alternative.
👍
Would be interesting to replace audio output pentodes with FETs. EL84 is not rare, but some other tubes like ECL82, ELL80 might be rare.
I’ve seen someone else do it before. Can’t remember where, but you can probably find it in RUclips.
Please please. Try to do a section on replacing EL84 with solid state. The tubes are expensive and I restore ideally and cannot add the cost of these tubes to my restoration. The N-mosfet should work just as well. Just the calculation in e.g. a philips radio.
I’ll probably have a go at that. These FETs won’t work because they can’t handle the power, but I’m sure something can be worked out.
@@electronicsoldandnew 💗
@@electronicsoldandnewA power FET attached into a heatsink probably is a way to go?
Wonder if a tube-shaped “hood” could be made of aluminium to act as a heatsink? That’s an idea.
🙂 nice!
👍
30:58 I was a little confused if this was AC or DC
Where I wouldn't mind outright replacing tubes is at the AC-to-DC stage. A lot of power could be saved there without having any meaningful effect on the circuit downstream of it, unless you are deliberately using power sag as a source of distortion like Eddie Van Halen did.
Agreed. The rectifier tube is one of the most common replacements, and power sag … you need deafening levels to achieve that. My ears wouldn’t handle it 😊
@@electronicsoldandnew Wrap it in a transparent case so it resembles a tube, and stick a little neon bulb in there to confirm B+ is good, and the only difference most people will notice is that it seems to warm up instantly.
If you use a series resistor, you’d actually achieve the sag effect, depending on the amp stage type (class A, or push-pull).
Good fun
👍
Hi Manuel, this video made me decide to write you about my recent idea, inspired by the problem i faced with few of my stereo radios, they need magic eye emm803, which is hard to find and very expensive. I made a simple circuit which i tested and working fine consisted of em84(6E2) green led and few components. I wanted to ask you if you agree to develop a pcb that can be fited into this chinese socket, so em84 goes into the socket and socket goes to the original socket of the radio without messing any components in the radio itself. I will email you a picture of the schematic and the final tube i made... so i look forward to hear what you think. Greetings from Bulgaria.
That sound interesting. I’d love to give it a try. Email me the details. By the way, if I do the video, others will be able to order their own PCBs with no benefit to you or me, so I hope you don’t mind. Contrary to popular belief, I don’t get paid commissions on PCBs ordered from my sponsor. I do get my own for free and they sponsor the videos, but that’s about it. I don’t want to make a business of this. It’s too much fun as a hobby 😊
@@electronicsoldandnew it's a hobby for me too, no problem, if you succeed to figure it out, i will order some of these pcbs👍
a question ive allwas wondered with detection diodes is why dont designers give them a little bit of pre bias to help them out ,something low enough to not have any ambient temperature effects something in the ballpark of 200mv for silicon 100mv for germainium so still cut of by a big margin but a bit more ready for signal,
i want to find out for myself right now but my "man lab" is currently repurposed as a bedroom grrr!
That is a good way to make them instantly sensitive, as is done with transistors. Should be easy in theory, but not so simple in practice se there are oppositely biased diodes, so opposing supplies would need to be designed, and could make the circuit a little messy.
@@electronicsoldandnew ah yes makes sense, thanks
Interesting!
There is a slight différence between the EABC 80 and a FET : the FET behaves like a pentode!
Sure, but as a proof of concept, it works 😊
Maybe it is time for a 12AX7 semiconductor replacement in both audio sweep/pink noise spectrum analysis and blindfold listening test in a guitar amp or stereo console, with respect to the tube.
That’s a challenge I won’t make to my tube amp enthusiasts 😊 I’d lose a lot of friends 😊
nice!