You are not the only one who has had problems with 833a tubes. I remember watching another video where the builder came to the conclusion that the tube was not very stable at higher HF frequencies. If may have been ElPaso TubeAmps, but I'm not sure. It has been awhile since I saw that video.
excellent. Got a few homebrew amps am playing with have u a contact facility to send a few quick vids to see what u think.. would love your opinion one a couple of ideas
I think you did an excellent job. Granted I'm not technically knowledgeable when it comes to building amps but I know when one builds something worthy of a compliment. How much do you think in parts you spent (minis tube cost and time) on the build. Advice: Join WWDX. The knowledge of many of the members on that particular site is incredible! Once again I think you did an excellent job. 73
A lot of the parts I already had.. the aluminium was quite pricey also the had to buy one of the vac caps. Probably total cost £800 or there about. Still, it’s cheaper than a commercial amp and looks a lot better
@@thewtf ok thanks. I'm trying to learn bias right now. I was low on power and had people tell me to increase the bias voltage to 18v..... I did that and I got less power. Not sure if I was being trolled or not. My tubes aren't running away. I have Penta that requires zero bias. But at zero bais my idle current is 400 ma, higher than I'd like plus it achieves no additional output for me compared to the 3.5v of bias at 200ma of plate current. With 200 watts of drive I can get 1800 watts with 300 watts I can get 1800 watts. Always remain above 3,000 vdc on the plate. So I typically don't drive with more than 175 watts. I can remove one tube and get 1100 on one at a time but only 1800 with two. Again always staying above 3000 vdc with one or two tubes. But I have noticed with two tubes I'm only getting 4.5 volts on the filaments when driving two tubes. 5 volts when only driving one. I'm not sure if the filament voltage would cause the loss in power or not. May need to buy a new transformer Hammond 256x5 is a good replacement. Just doesn't have a center tap.
Got a question for you Justin, re the input/output relay at the back of the RF deck... Is it a simple DPDT type that bypasses the amplifier once de-engerised? Do you have any problems with RFI due to the input and output connectors being on the RF deck?
@@thewtf thanks for the information, Justin. I thought I was going to have to fit 2 separate relays, one on the input and one on the output, according to conflicting opinions...
For 3000 volts on the plate it should be putting out around 1200 watts bring the bias down to what it should be then work on the problem what is causing the high idle current.
@@thewtf I am assuming you are blowing fuses once you apply RF but it is OK on DC key down with no RF. If this is the case you have a problem with your plate choke and RF getting into the power supply. I worked with many of these graphite tubes and the DC condition with 7.5v at 3KV approx. is pretty much the same in all cases. I would be investigating series resonance with the plate choke as that will also affect your plate tuning circuit on 20m. I am curious why you are running plate coupling caps in series, I am assuming here you have too higher capacitance and need to drop the capacitance, 1000 - 2200pf even up to 3300pf would suffice, if you wanted to beef up power handling of the plate capacitors they should be in parallel however these caps you are using look pretty chunky and one would suffice. Another RF ground improvement you can do is run a common copper strap to all of your RF connections as you are relying on the alloy chassis and may be getting issues with resistance causing the stability problems and even vhf oscillation. You should also focus on cooing the tube bases as they get very hot, so blowing air from underneath and fabricating some sort of chimney (I have used perspex tube in the past) would be an improvement over the current cooling, make sure your the chimney system can direct air onto the anode or make a large anode cooler as well. Great video to watch, well done on the home brewing. 73 Brendan VK3MH
@@vk3mh Hi Brendan, thanks for the comments. Yes, the plate coupling caps were a bit high hence the reason thy are in series. I had thought about series resonance in the plate choke and did check for this using the GDO but it seems OK; I did even replace it. The problem occurs only if the HT applied is > 3kv..at the moment I run it at 2.8 kv and it works fine...can get a kilowatt out on all the bands so can't really complain. There's nothing really to suggest there's parasitic oscillation and again it is something I have thought about. Yes, the cooling does actually blow air on to the bases and the glass envelopes as the fan is mounted split level blowing air directly on to the bases. Been using it fairly regularly and it has been pretty good. The only problem I have had is the HT lead connection in the ceramic insulator came adrift so was not getting any volts. That's all been sorted! Keep watching the videos, got another project in the pipeline. 73's Justin GW0FZY
Always fun building it, then the switch on moment :) and all the fun too be had after, built many of em small and big, if ya follow my youtube name you might find some of em that i uploaded, albeit im no video specialist , .good build, lots of space for another one, at 3k they give good power 1kw+ and at 160m it's needed to work from South Africa to G country.. im not active anymore but still have all the gear... maybe someday i'll return zs2cx
Just stumbled upon the video again, and the parasitic chokes caught my eye, i once ran into similar problems with tha 4-1000a tubes, the parasitic choke changed to hairpin did the trick. the 3-500z i built had 7.5v (diodes) at 3500v, the 4-1000a single tube had 9v (diodes) when keyed.. 2 3500z at 2.8kv should easy do 1kw, and with more drive 1.5kw. just checked on mine a day or so ago, i found the blue peppermint caps to be awfull(heating up...).. changed the 3500z amps out to very old brown Dielectric type, works like a charm, i found the decoupling caps (blue peppermint) to be a problem as well on 20m. i did not catch your ht voltage ? maybe screen the filament choke a bit as well you might have some stray pfft in there. 1 thing at a time , i know :) one other note : the chinese tubes want a little bit more filament voltage .5v higher, and then they run like hell! anyway hope it gives you lots of joy, im not on hf, yet my towers are still up and all the gear still in ham room :) ... maybe next year, i will run on TB again and catch you there. Johan zs2cx
@The WTF ok. No chimney, no sunken sockets. Etc. No, I didn't see a fan, but where-ever it is, it's not enough. Not sure why, because otherwise it's a fine build.
@@yakpilot9596 the proof of the pudding is in the eating… yes I agree the purist would require the things you mentioned; its an amateur radio amp and it works 100%
I'm not a "purist" by any means, but I have built several 3-500 tube amps, and have operated a lot of Henry amps that use them as well. That being said, I also read the spec sheet on these tubes that require the base seal to be kept below 200C and the plate seal below 225C. Thus the need for forced air cooling (typically) through depressed sockets and usually chimneys as well. Of course it works now, and it will do so right up until it doesn't. I've done exactly the same thing you did, and learned how to solder filament pins in the process thus my tongue in cheek comment. It's a beautiful amp. It is well done. If you run that pair around 800 mils and you get the anodes bright red, or even cherry red, you'll remember this respectful input.
@@yakpilot9596 I totally agree. I think if it was ever run in CCS spec then you're quite correct. I don't run it anywhere near full chat so I guess that's probably why it has not given any trouble. The layout turned out the way it did as it did not start out as a 3-500z amp but was originally intended for a pair of Gu48 triodes. Thanks for the comments.
Fantastic, made my jaw drop, some lovely workmanship and great to see valves in use. Thanks for the upload.
Best regards from Carl,G0lka.
Jaw dropped. Just wow. Love it. I just checked google satellite maps and we can see you at night. The glow.
3-500Z are good tubes, i like how they glow orange like it has a ghost possession, i have those in my 51 year old drake 2kw L4B amplifier
Very nice description and demonstration.
ground each tube grid pin with individual copper strap.
Gosh, that box is huge you could make a 3CX10.000 arrangement in that. nice work though.
You are not the only one who has had problems with 833a tubes. I remember watching another video where the builder came to the conclusion that the tube was not very stable at higher HF frequencies. If may have been ElPaso TubeAmps, but I'm not sure. It has been awhile since I saw that video.
Very neat work Justin.
The metal handles for the cabinet, can I ask where did you get them from, please? 🤔
Scrapped old equipment
Hi Justin, someone elsewhere asked where was the cooling fan? 🤔 Good question. 😉
Its on the side opposite the valves!
Looks well built
RUclips should require every video to include at least one cat.
I have ssb 100 watt trasmitter .can i connect with linier of vacuum? How to calculate capasitor coupling. Or how your method or suggestion.🙏thanks
Bias is too high, should be +10V on the cathodes with 3KV anode voltage, the parasitic chokes look a bit strange, how did you design them?
excellent. Got a few homebrew amps am playing with have u a contact facility to send a few quick vids to see what u think.. would love your opinion one a couple of ideas
I think you did an excellent job. Granted I'm not technically knowledgeable when it comes to building amps but I know when one builds something worthy of a compliment. How much do you think in parts you spent (minis tube cost and time) on the build. Advice: Join WWDX. The knowledge of many of the members on that particular site is incredible! Once again I think you did an excellent job. 73
A lot of the parts I already had.. the aluminium was quite pricey also the had to buy one of the vac caps. Probably total cost £800 or there about. Still, it’s cheaper than a commercial amp and looks a lot better
have a home brew amp with same tube that a friend cannot get to function. Fancy taking a look at a vid i could send over for your thoughts ?
Nice and clean build
Hi OM. Really pretty build. 73 W5DN
So u had to increase the bias voltage to reduce the plate current? Doesn't that also reduce the power output?
Yes it would
@@thewtf ok thanks. I'm trying to learn bias right now. I was low on power and had people tell me to increase the bias voltage to 18v..... I did that and I got less power. Not sure if I was being trolled or not. My tubes aren't running away. I have Penta that requires zero bias. But at zero bais my idle current is 400 ma, higher than I'd like plus it achieves no additional output for me compared to the 3.5v of bias at 200ma of plate current. With 200 watts of drive I can get 1800 watts with 300 watts I can get 1800 watts. Always remain above 3,000 vdc on the plate. So I typically don't drive with more than 175 watts. I can remove one tube and get 1100 on one at a time but only 1800 with two. Again always staying above 3000 vdc with one or two tubes. But I have noticed with two tubes I'm only getting 4.5 volts on the filaments when driving two tubes. 5 volts when only driving one. I'm not sure if the filament voltage would cause the loss in power or not. May need to buy a new transformer
Hammond 256x5 is a good replacement. Just doesn't have a center tap.
so sick love the tube glow
saludosamigo tengo una valvula 3-500z para construir un lineal para fm me gustaria ver un plano para construirlo o que me orientes en su construccion
Looks nice but 4.8-5.2 volts on the bias.. try 8 1000v 3 amp diodes to ground 👌🏻
Any books on that?
Got a question for you Justin, re the input/output relay at the back of the RF deck...
Is it a simple DPDT type that bypasses the amplifier once de-engerised? Do you have any problems with RFI due to the input and output connectors being on the RF deck?
Yes it is a DPDT ... I haven’t had any problems with RFI and usually put the changeover relay on the RF deck on the linears I have made
@@thewtf thanks for the information, Justin. I thought I was going to have to fit 2 separate relays, one on the input and one on the output, according to conflicting opinions...
Hi how much voltage is needed for 4 of them 3-500z and can i used 3-400z with 3-500z on a emergency just a curiosity... thanks for the time 73
About 2.8-3kv but more current +\- 1 amp
@@thewtf thanks 🙏 73
Can I mix two kind of tubes 3-400z with 3-500z. Say two of each on one box .... just a question...
Bom dia você e de onde
Question are you not worried about x-rays ?
Not at these voltages/currents
For 3000 volts on the plate it should be putting out around 1200 watts bring the bias down to what it should be then work on the problem what is causing the high idle current.
Thanks. Yes, I did all that but couldn’t stop it from blowing the fuse! It will give a kilowatt at the moment which is plenty
@@thewtf I am assuming you are blowing fuses once you apply RF but it is OK on DC key down with no RF. If this is the case you have a problem with your plate choke and RF getting into the power supply. I worked with many of these graphite tubes and the DC condition with 7.5v at 3KV approx. is pretty much the same in all cases. I would be investigating series resonance with the plate choke as that will also affect your plate tuning circuit on 20m. I am curious why you are running plate coupling caps in series, I am assuming here you have too higher capacitance and need to drop the capacitance, 1000 - 2200pf even up to 3300pf would suffice, if you wanted to beef up power handling of the plate capacitors they should be in parallel however these caps you are using look pretty chunky and one would suffice. Another RF ground improvement you can do is run a common copper strap to all of your RF connections as you are relying on the alloy chassis and may be getting issues with resistance causing the stability problems and even vhf oscillation. You should also focus on cooing the tube bases as they get very hot, so blowing air from underneath and fabricating some sort of chimney (I have used perspex tube in the past) would be an improvement over the current cooling, make sure your the chimney system can direct air onto the anode or make a large anode cooler as well. Great video to watch, well done on the home brewing. 73 Brendan VK3MH
@@vk3mh Hi Brendan, thanks for the comments. Yes, the plate coupling caps were a bit high hence the reason thy are in series. I had thought about series resonance in the plate choke and did check for this using the GDO but it seems OK; I did even replace it. The problem occurs only if the HT applied is > 3kv..at the moment I run it at 2.8 kv and it works fine...can get a kilowatt out on all the bands so can't really complain. There's nothing really to suggest there's parasitic oscillation and again it is something I have thought about. Yes, the cooling does actually blow air on to the bases and the glass envelopes as the fan is mounted split level blowing air directly on to the bases. Been using it fairly regularly and it has been pretty good. The only problem I have had is the HT lead connection in the ceramic insulator came adrift so was not getting any volts. That's all been sorted! Keep watching the videos, got another project in the pipeline. 73's Justin GW0FZY
Always fun building it, then the switch on moment :) and all the fun too be had after, built many of em small and big, if ya follow my youtube name you might find some of em that i uploaded, albeit im no video specialist , .good build, lots of space for another one, at 3k they give good power 1kw+ and at 160m it's needed to work from South Africa to G country.. im not active anymore but still have all the gear... maybe someday i'll return
zs2cx
FB OM de GW0FZY ( ex ZS6BLU)
Just stumbled upon the video again, and the parasitic chokes caught my eye, i once ran into similar problems with tha 4-1000a tubes, the parasitic choke changed to hairpin did the trick. the 3-500z i built had 7.5v (diodes) at 3500v, the 4-1000a single tube had 9v (diodes) when keyed.. 2 3500z at 2.8kv should easy do 1kw, and with more drive 1.5kw. just checked on mine a day or so ago, i found the blue peppermint caps to be awfull(heating up...).. changed the 3500z amps out to very old brown Dielectric type, works like a charm, i found the decoupling caps (blue peppermint) to be a problem as well on 20m.
i did not catch your ht voltage ? maybe screen the filament choke a bit as well you might have some stray pfft in there.
1 thing at a time , i know :)
one other note : the chinese tubes want a little bit more filament voltage .5v higher, and then they run like hell!
anyway hope it gives you lots of joy, im not on hf, yet my towers are still up and all the gear still in ham room :) ... maybe next year, i will run on TB again and catch you there.
Johan zs2cx
Hi, subscribed.
What is this?! Amplifier?... Of audio hifi?! Voltage amp??? What, Interesting...
Rf amplifier. Will do a big audio amp soon..keep watching!!
Not bad ! but you have a big box, so add 2 more lamps and at least one!
Vy73 and I se You on the bands :-)
Yes the box is quite big ... was supposed to have 2x 833a’s
Get ready to learn how to resolder filament pins. I.E. No tube cooling.
There is a fan you just didn’t spot it
@The WTF ok. No chimney, no sunken sockets. Etc. No, I didn't see a fan, but where-ever it is, it's not enough. Not sure why, because otherwise it's a fine build.
@@yakpilot9596 the proof of the pudding is in the eating… yes I agree the purist would require the things you mentioned; its an amateur radio amp and it works 100%
I'm not a "purist" by any means, but I have built several 3-500 tube amps, and have operated a lot of Henry amps that use them as well. That being said, I also read the spec sheet on these tubes that require the base seal to be kept below 200C and the plate seal below 225C. Thus the need for forced air cooling (typically) through depressed sockets and usually chimneys as well.
Of course it works now, and it will do so right up until it doesn't. I've done exactly the same thing you did, and learned how to solder filament pins in the process thus my tongue in cheek comment.
It's a beautiful amp. It is well done. If you run that pair around 800 mils and you get the anodes bright red, or even cherry red, you'll remember this respectful input.
@@yakpilot9596 I totally agree. I think if it was ever run in CCS spec then you're quite correct. I don't run it anywhere near full chat so I guess that's probably why it has not given any trouble. The layout turned out the way it did as it did not start out as a 3-500z amp but was originally intended for a pair of Gu48 triodes. Thanks for the comments.
Nice fun but my pentruder wallsaw makes me money 480v 400hz
NICE W4GSM
Order a Harback input board
Change the jelorics to a horseshoe configuration
art looks like its from doctor whos tardis
Nowt wrong with conventional !!! 🙃😉
i like this de YB9AGA
Super! G4NGV
Have a look at mr bbi
Justin BBI is really good , on tube & pill Amp’s . JB Utah ( KJ7TBR)
Got enough room for 4 1000
How much to build a 4 3-500z