Found it www.g7syw.com/download/Cobra_200GTL-DX_Service_Manual.pdf Interestingly it uses the RD16HHF1 mosfet as final. Edit: my mistake - it uses a pair of 2SC2290s...
I use an etching process with this "fab in a box" transfer paper. pcbfx.com/main_site/pages/start_here/overview.html I did a video on the process a while back here ruclips.net/video/X5O1au3FvRc/видео.html - the process description starts around the 04:00 mark. I was using Eagle back then but it works exactly the same for kicad. Its a bit of fiddling around but the results are usually OK. You can even use the same approach for double sided boards too. Thank you for the comment!
thanks for the video, would really like a video of the process you'd take to design from scratch an RF Power Amp in the range of 50 -100W in the HF Band..there are videos on here but everyone blows through the actual design process and they make it really complicated.. was just wondering if you could ?
50-100W might be a bit beyond my capabilities ;) - let me have a think though. If you are interested in the math I would recommend this video from Charlie Morris - I think he does a great job explaining each stage - and the matching process for each stage: ruclips.net/video/eEuDxwP6CpA/видео.html There's a lot to the design process but simply put you start with the final stage and from your desired power determine the load resistance the formula VCC^2/2*power. You then work backwards and at each stage calculate a new power which is 10-15 dBs below the following stage. This gives you a new load resistance which you then have to match to the input impedance of the following stage (typically with a toroid transformer - but there are a vast array of matching techniques). Thats for a class A amplifier anyway - its such a deep topic Another good resource is Solid State design for the radio amateur - Chapter 4. I think its out of print but you can pick up used copies on Amazon. There's also archive.org that has an eVersion. I probably couldn't do a better job than Charlie did but let me hit the books and see what I can come up with. Thank you very much for the comment
hehe - he isn't wrong on that. A simple example is calculating the input impedance of that final IRF510 stage. There's no doubt it can be done from first principles - I can't though and I had to resort to modelling it in LTSpice and using that. Then you get variations in the components themselves for example those eBay bought 2SC5706s - that had me stumped for quite a while. The component tester I had showed them being identical but they behaved quite differently in circuit. Stay tuned - I will definitely be doing some more playing around...
@@na5y I think a lot of the bogus parts are either (A) production runs that didn't meet specs at the fab or (B) counterfeits that are relabeled parts that can be had cheaper than the real ones. In the latter case, I think the counterfeiters get away with it a lot because in simple DC circuits the transistors don't get pushed to their real limits. So if you're just making a power supply or something, the differences in RF characteristics don't really show up much... very annoying.
Very nice work, the two driver stages probably accounts for reasonable output on 14MHz. I'd be interested to see you replace the IRF510 with a Mitsubishi RD16HHF1 and report the difference at 14MHz. I'd bet your IRF510 amp would go gangbusters at 7 to 10MHz. PS: 'bodge' shows your antipodean origins!
Thank you Paul - guilty as charged on coming from Australia, living in Dallas area now but grew up in Newcastle, NSW. I might have to try a 7Mhz version and post results! I'll have to check out that other mosfet too - datasheet seems to put it somewhere between the IRF510 and IRF530
@@na5y The IRF510 never shines in an RF role, in the first place, but 14MHz is beyond its sweet spot even in the RF misuse case :-). I know there are people who have successfully pushed it as far as the 10m band, but those are real accomplishments, not "normal" performance. In my experiments, I don't really like what the IRF510 does above about 7MHz. I'm surprised you're only getting 3W out of that arrangement... At full swing on the drain, an 8-ohm load to the IRF510 with a 12V supply a proper transistor should be giving you about 8W. It's likely the IRF510's high gate capacitance and weakness WRT to rise time and reverse recovery time :-(.
You've given me impetus to try again at 7MHz. 3.6 W was about all I could get out of it before it went non-linear. Thank you for the comment Yakov. Is there an alternative to the IRF510 you could suggest?
@@na5y I've been playing with the 2SK3476 and working with a genuine RF transistor is like breathing fresh air. It's not the most convenient package... in my case, I was pleasantly surprised to have no oscillation issues to fight, though a home etched board without thermal vias to a ground plane might require some TLC :-). I got a board from OSHPark with a proper footprint. I've been thinking of just trying one dead-bug style, glued to a heat sink, though... I suspect it would work, but don't know for sure. I think IRF510s are popular because they're easy to heatsink and hams are super frugal...
Very nice. Clean layout. Keep posting n happy summer greetings
Thank you - and believe me it really felt like summer in that garage!
Enjoyed the video, many thanks for sharing!
Thank you Alex - Glad you enjoyed!
very nice construction with SMD components too :-)
Thank you Amir - I enjoy doing work with SMD components. There's nothing smaller than 0805 on the board and It does help to have a hot air gun!
What about copying the power output stage of the COBRA 200 GTL-DX
Is there a schematic around - I couldn't find one
@@na5y what about the COBRA 200 GTL-DX Service Manual.
I'll have a look there - thank you for the comment
Found it www.g7syw.com/download/Cobra_200GTL-DX_Service_Manual.pdf Interestingly it uses the RD16HHF1 mosfet as final. Edit: my mistake - it uses a pair of 2SC2290s...
Appreciated, thank you. What technique do you use to cut the neat lines in the circuit board?
I use an etching process with this "fab in a box" transfer paper.
pcbfx.com/main_site/pages/start_here/overview.html
I did a video on the process a while back here ruclips.net/video/X5O1au3FvRc/видео.html - the process description starts around the 04:00 mark. I was using Eagle back then but it works exactly the same for kicad. Its a bit of fiddling around but the results are usually OK. You can even use the same approach for double sided boards too.
Thank you for the comment!
thanks for the video, would really like a video of the process you'd take to design from scratch an RF Power Amp in the range of 50 -100W in the HF Band..there are videos on here but everyone blows through the actual design process and they make it really complicated.. was just wondering if you could ?
50-100W might be a bit beyond my capabilities ;) - let me have a think though. If you are interested in the math I would recommend this video from Charlie Morris - I think he does a great job explaining each stage - and the matching process for each stage:
ruclips.net/video/eEuDxwP6CpA/видео.html
There's a lot to the design process but simply put you start with the final stage and from your desired power determine the load resistance the formula VCC^2/2*power. You then work backwards and at each stage calculate a new power which is 10-15 dBs below the following stage. This gives you a new load resistance which you then have to match to the input impedance of the following stage (typically with a toroid transformer - but there are a vast array of matching techniques). Thats for a class A amplifier anyway - its such a deep topic
Another good resource is Solid State design for the radio amateur - Chapter 4. I think its out of print but you can pick up used copies on Amazon. There's also archive.org that has an eVersion.
I probably couldn't do a better job than Charlie did but let me hit the books and see what I can come up with. Thank you very much for the comment
@@na5y yeah i'm aware of charlie and love his content, but charlie says PA are like a black art .
hehe - he isn't wrong on that. A simple example is calculating the input impedance of that final IRF510 stage. There's no doubt it can be done from first principles - I can't though and I had to resort to modelling it in LTSpice and using that. Then you get variations in the components themselves for example those eBay bought 2SC5706s - that had me stumped for quite a while. The component tester I had showed them being identical but they behaved quite differently in circuit. Stay tuned - I will definitely be doing some more playing around...
@@na5y I think a lot of the bogus parts are either (A) production runs that didn't meet specs at the fab or (B) counterfeits that are relabeled parts that can be had cheaper than the real ones. In the latter case, I think the counterfeiters get away with it a lot because in simple DC circuits the transistors don't get pushed to their real limits. So if you're just making a power supply or something, the differences in RF characteristics don't really show up much... very annoying.
Very nice work, the two driver stages probably accounts for reasonable output on 14MHz. I'd be interested to see you replace the IRF510 with a Mitsubishi RD16HHF1 and report the difference at 14MHz. I'd bet your IRF510 amp would go gangbusters at 7 to 10MHz. PS: 'bodge' shows your antipodean origins!
Thank you Paul - guilty as charged on coming from Australia, living in Dallas area now but grew up in Newcastle, NSW. I might have to try a 7Mhz version and post results! I'll have to check out that other mosfet too - datasheet seems to put it somewhere between the IRF510 and IRF530
@@na5y The IRF510 never shines in an RF role, in the first place, but 14MHz is beyond its sweet spot even in the RF misuse case :-). I know there are people who have successfully pushed it as far as the 10m band, but those are real accomplishments, not "normal" performance. In my experiments, I don't really like what the IRF510 does above about 7MHz. I'm surprised you're only getting 3W out of that arrangement... At full swing on the drain, an 8-ohm load to the IRF510 with a 12V supply a proper transistor should be giving you about 8W. It's likely the IRF510's high gate capacitance and weakness WRT to rise time and reverse recovery time :-(.
You've given me impetus to try again at 7MHz. 3.6 W was about all I could get out of it before it went non-linear. Thank you for the comment Yakov. Is there an alternative to the IRF510 you could suggest?
@@na5y I've been playing with the 2SK3476 and working with a genuine RF transistor is like breathing fresh air. It's not the most convenient package... in my case, I was pleasantly surprised to have no oscillation issues to fight, though a home etched board without thermal vias to a ground plane might require some TLC :-). I got a board from OSHPark with a proper footprint. I've been thinking of just trying one dead-bug style, glued to a heat sink, though... I suspect it would work, but don't know for sure. I think IRF510s are popular because they're easy to heatsink and hams are super frugal...
That is an interesting package toshiba.semicon-storage.com/ap-en/semiconductor/product/mosfets/detail.2SK3476.html . 6 bucks a pop though!