Video looking at the IRF510 based push pull RF power amplifier. The video covers board construction, IRF510 quiescent current adjustment and final output tests.
Thanks very much. Its really the efficient way to avoid your MOSFETs from burning out as well as to ensure the circuit board contains to shorted tracks. I would request if a tutorial could be made on transformer coil winding.
That's what I was thinking too Stephen. My plan was to swap out the trim pots tonight and keep playing. It will make bias adjustment so much easier! Charlie
Stephen. I replaced the pots with 10 turn versions, which made adjusting the quiescent current easier. I then played around with the bias while watching the output. It did make a small difference, but didn't totally remove the distortion. I also played around with setting the quiescent current to 70mA. No major difference. I've now return it to 50mA. All good fun! 73s Charlie
Nice board Charlie. I might try duplicating one of these for my mighty mite, (to make it mightier.) What is the value of the bias resistors, so I can order in some 10 turn versions.?
Hi Fanima. The original single turn pot was a '104' or 100k, as I couldn't find any 10k ones. The 10 turn pots that are now in the circuit are 10k. At the end of the day it's a voltage divider so 10+ k ohm will be fine. Anything less than 10k and the current through the pot starts to increase and you risk burning it out. I thought the dremel went well. I tried to make up a small jig so i could slide the board and keep the dremel still. That way the straight lines were straighter than by hand.
Enjoyed ,Just finished watching the set up ,good idea using the resistor measuring the volts across it ,i have pinged current meters in the past ,so will try this ,Always learning at my age hi hi Thanks again Charlie ,next is i will look at noting down ur low pass filter.
By having separate bias adjustments for each IRF510, matching the devices probably isn't critical. You could check this by looking for distortion on the output with a scope, if the upper and lower portions of the waveform are equal, then the devices are probably close in gain. If the bias is too far off you might see crossover distortion. BTW you might try putting IRF510's in push-pull parallel. I did see a design that used 8 of them in that configuration, with a claimed KW output on 6 meters!
The newer IRF types are usually even more designed for switching. Older batches behave a little better in linear applications. When adjusting the quiescent current, I found that this "kink" is softer. But even today the IRFs are still different. To optimize it is still worthwhile to measure a few. By the way, your voice sounds like a cold. Get well! 73 Bernd
Thanks Bernd. In line with Stephen's suggestion I do want to try and match a couple, but unfortunately, I only had two in the shack. I'm looking to order a bunch and keep playing. Yes, I've had a cold for a week, which is a real pain. I have an international flight tomorrow, so I'm hoping I'm a bit better by then. I don't want to be 'that guy' coughing all flight!
Hi Paul. Just did rough test. I first set the sig gen to 3MHz at the set the output to get 10W out. I then increased the sig gen freq and recorded the PA output on the watt meter and oscope. 3MHz = 10W, 67Vp=p, 7MHz = 4.5W, 43Vp-p, 14MHz = 1W, 20Vp-p, 18MHz = 0.5W, 14Vp-p. Charlie
@@CharlieMorrisZL2CTM Just as the uBitx folks have found. Still a fine QRP amp for lower HF. WA2EBY claimed usable output power above 10mhz but he ran his amp at 28 volts (from memory).
That's right. 28VDC seems to be a common voltage. I was looking to use a 19VDC laptop powerpack, but it was limited to 2A or so. I will look at that again at some stage. Charlie
12v power with FET rf amps is a problem. 'EBY did claim close to 10w a 12v on ten meters though. 28v is pushing it, but you should get decent performance at 18-20 volts. That 2A supply should work, provided it can handle that load key-down.
Hi Paul. I'm currently travelling (again...) and don't have any notes with me. Given that it was a base rig, it would have been 13.8VDC. The small portable rigs are all 12VDC given the battery pack I have.Charlie
This is what ham radio is supposed to be all about. Thank you. 73
Thanks. An no IRF510s were harmed in the making of the video. Even better!
73s
Charlie
Hi Charlie! I assembled and checked your circuit works fine!
That's great Vladimir.
Thanks very much. Its really the efficient way to avoid your MOSFETs from burning out as well as to ensure the circuit board contains to shorted tracks. I would request if a tutorial could be made on transformer coil winding.
Good job there, Charlie. Thanks. I wonder if, with the 10-turn pots, tweaking the bias voltages could help with output waveform shape.
That's what I was thinking too Stephen. My plan was to swap out the trim pots tonight and keep playing. It will make bias adjustment so much easier!
Charlie
Stephen. I replaced the pots with 10 turn versions, which made adjusting the quiescent current easier. I then played around with the bias while watching the output. It did make a small difference, but didn't totally remove the distortion. I also played around with setting the quiescent current to 70mA. No major difference. I've now return it to 50mA. All good fun!
73s
Charlie
Nice board Charlie. I might try duplicating one of these for my mighty mite, (to make it mightier.) What is the value of the bias resistors, so I can order in some 10 turn versions.?
Hi Fanima. The original single turn pot was a '104' or 100k, as I couldn't find any 10k ones. The 10 turn pots that are now in the circuit are 10k. At the end of the day it's a voltage divider so 10+ k ohm will be fine. Anything less than 10k and the current through the pot starts to increase and you risk burning it out. I thought the dremel went well. I tried to make up a small jig so i could slide the board and keep the dremel still. That way the straight lines were straighter than by hand.
Enjoyed ,Just finished watching the set up ,good idea using the resistor measuring the volts across it ,i have pinged current meters in the past ,so will try this ,Always learning at my age hi hi Thanks again Charlie ,next is i will look at noting down ur low pass filter.
Great. I also need to do some more with PA design. Hopefully, I'll get a chance this summer.Charlie
By having separate bias adjustments for each IRF510, matching the devices probably isn't critical. You could check this by looking for distortion on the output with a scope, if the upper and lower portions of the waveform are equal, then the devices are probably close in gain. If the bias is too far off you might see crossover distortion.
BTW you might try putting IRF510's in push-pull parallel. I did see a design that used 8 of them in that configuration, with a claimed KW output on 6 meters!
Thanks for the feedback. At this stage I'm quite happy with the performance. Not bad for so few components.
The newer IRF types are usually even more designed for switching. Older batches behave a little better in linear applications. When adjusting the quiescent current, I found that this "kink" is softer. But even today the IRFs are still different. To optimize it is still worthwhile to measure a few.
By the way, your voice sounds like a cold.
Get well!
73 Bernd
Thanks Bernd. In line with Stephen's suggestion I do want to try and match a couple, but unfortunately, I only had two in the shack. I'm looking to order a bunch and keep playing.
Yes, I've had a cold for a week, which is a real pain. I have an international flight tomorrow, so I'm hoping I'm a bit better by then. I don't want to be 'that guy' coughing all flight!
Great instructional video Charlie. These FETs drop off above 10mhz, would interest me to see by how much, at 14 and 18mhz.
Hi Paul. Just did rough test. I first set the sig gen to 3MHz at the set the output to get 10W out. I then increased the sig gen freq and recorded the PA output on the watt meter and oscope. 3MHz = 10W, 67Vp=p, 7MHz = 4.5W, 43Vp-p, 14MHz = 1W, 20Vp-p, 18MHz = 0.5W, 14Vp-p.
Charlie
I should say too Paul that there is a 1dB drop in sig gen output from 10MHz to 18MHz. You'll need to factor that into the results above.
@@CharlieMorrisZL2CTM Just as the uBitx folks have found. Still a fine QRP amp for lower HF. WA2EBY claimed usable output power above 10mhz but he ran his amp at 28 volts (from memory).
That's right. 28VDC seems to be a common voltage. I was looking to use a 19VDC laptop powerpack, but it was limited to 2A or so. I will look at that again at some stage.
Charlie
12v power with FET rf amps is a problem. 'EBY did claim close to 10w a 12v on ten meters though. 28v is pushing it, but you should get decent performance at 18-20 volts. That 2A supply should work, provided it can handle that load key-down.
can the mosfet be replaced with irf530 friends
Hi Charlie Power Amplifier irf510. 50ma -I drain?
what input power did you use?
What was the supply voltage for this test Charlie? 12v, 13.8v?
Hi Paul. I'm currently travelling (again...) and don't have any notes with me. Given that it was a base rig, it would have been 13.8VDC. The small portable rigs are all 12VDC given the battery pack I have.Charlie
Charlie, you gave up on zener diodes to gate s hi😁
They are there Mikele. Just to the side of the two 27ohm resistors. I didn't see any negative impact, which is good.
Charlie
@@CharlieMorrisZL2CTM it s all great in schematics,no problems but you have to be careful where you buy fets 😉
Thanks Mikele. I'm looking to use Kits and Parts for the purchase. I don't trust my normal source, even if they are cheap!
Jameco, Mouser and Digikey should be safe sources for FETS. Don't know if Vishay, IR, On Semi (Fairchild), or STM are the best choice though.
Great. Thanks for that. I'll try Kits and Parts as I need some toroids too.
Ok. PCB
Ok