And Rick immediately gets added to my Christmas Card list.
5 Likes
So, I went a different direction. I was too impatient to order and wait for a 10k SMD resister, BUT I have a bunch of 1/4 watt through-hole resistors in my inventory, so, I experimented with different resistor values across R35 because itās physically larger and therefore easier to work with. I soldered a 33 ohm resistor across it and have now gotten the amp to behave acceptably. I think both of our approaches get to the same outcome, which is to bring the detected reverse voltage closer to ground, although if I were to recommend one approach, I think I would recommend yours as mine kind of introduces an imbalance in a spot where thereās more reactive current flowing and is therefore probably more likely to let the smoke out of the amplifier sooner. So far so good, though. Iāve been getting good signal reports with it. The scope tells me about 40-50 watts PEP.
5 Likes
I went ahead and made a video detailing how I addressed the SWR issue. https://youtu.be/39iIoGDtQkA
4 Likes
Another update. I tried replacing the UHF connectors with BNC connectors both mounted to an aluminum plate I built. Everything worked fine with a pure resistive load, but then the amp went CRAZY with an antenna attached. Suspecting that the input and output are not meant to share a ground, thus the aluminum plate I had connecting the grounds of the input and output connectors was the problem, I took that mod out and replaced the UHF connectors. Itās still lost its mind. Still works into resistive loads, but now it cannot be hooked to any antenna in any configuration. It reads wildly inaccurate SWR and power levels. I obviously broke something, just not sure what I broke.
FWIW, the multimeter reads continuity on the circuit board between the ground of the input and output connectors, but Iām not sure if that DC continuity is also AC continuity. I still suspect I busted something by connecting the grounds on those two connectors. Or I clumsily overheated a component with my soldering iron or broke something manhandling the circuit board out of the enclosure. Endless list of possibilities for how I screwed this up.
If youāre going to replace the connectors with BNC, be careful!
2 Likes
Hi
Thanks for the warning, but done, and working just fine
I used the same idea of an internal ali mounting plate, secured by a central self-tapping screw to the back plate. I didnāt remove the board, just worked carefully to remove and reinstall the connectors.
One problem was that the BNCās I had to hand penetrated a little deep towards the board so there was a risk of the connector casing shorting to the pad on the board. I resolved this by padding the connector with a 10mm washer - the ones I had to hand happened to be copper brake hose crush washers.
Result looks ok ā¦
Rick
4 Likes
Thanks for sharing! Iāve got mine working now with R36 shorted to ground, therefore completely disabling the SWR protection. Iāll just have to be really careful with it now. Iāll reinstall my BNC connectors when I get home and test it out. Will post pictures of my implementation when Iām done. Iām starting to learn a lot about how this thing works just because Iāve spent so much time probing around in it with the oscilloscope. I found the schematic for the amplifier thatās at the core of this thing: IRF530 30w Mosfet Amplifier ā F5NPV ā V85NPV From the schematic, I can see that the input and output are meant to share a ground. So, obviously my SWR freakout is caused by some other defective component or something. Regardless, I feel like even if I blow up the finals, theyāre easy to replace, so Iām not too concerned. Looking forward to having BNC connectors on this.
2 Likes
Neat job.
It would be so much better if all low power (<400w) devices used BNC (in and out). I have BNC adaptors on all my gear, apart from the Elecraft kx3 and some vhf gear.
73 Andrew VK1DA/VK2DA
5 Likes
I couldnāt agree more. I donāt understand why UHF connectors persist. Theyāre inferior in almost every way!
3 Likes
Hi guys,
I confirm the SWR issue at my PA50 too. The problem starts only with a real antenna, with a dummy load all looks ok. Up to 20m all is ok, start with 17m up to 10m the SWR problem begun if I use a real (but resonant) antenna or my home antenna which is tuned by a CG3000 outdoor (SWR ever < 1.3 after tuning). I do the R36 mod too, but I havenāt a SMD resistor with 4k7 too 
I use a 10k through-hole resistor parallel to the R36. After this mod I can confirm, the shown SWR looks more correctly compared to an external SWR meter.
Here my pic for this mod (looks not very nice, but it works 
):
Later I will be change the R36 with a SMD 4k7 resistor.
73 Heiko, DL1BZ
3 Likes
Hi guys,
I tested something else and made comparisons with my external SWR meter. I think, 5k ist not enough at R36. More a value between 1k5ā¦2k2. I remove the R36 SMD resistor now completely and replace it with a 10k spiral potentiometer for better adjustment:
After this I repeat some SWR measurements and play around a little bit with the values. At the end I found 1k5 as best compromise. I have results near reality at 20m-10m and between, but at the low bands 80m-30m the shown SWR is a little bit to low than it should be, but not so much.
I use the PA50 more for portable so it is more important that the function is better above 10Mhz than lower.
In general I think, it would have been better if the protection was only effective from an SWR of 2.5 or 3 and not already start with only 2.0 . Iām sure, that would not be a problem for the finals, the most TRX do it so.
At the end I wouldnāt say āDonāt buy this amplifierā. We are hamradio operators and can solve such problems themselves. Aside from the SWR issue, itās a good amp I think and a good add-one for portable.
For more comfort I build with an ESP32 (TTGO T-DISPLAY) a small device, which makes a Blutooth connection between the ESP32 and my IC-705, get itās CAT data and switch the PA on/off with a 4N25 and generate an analog band switching signal for automatic switch the band filters in the PA (the IC-705 has no bandswitching output):
I setup the PA50 for XIEGU band switch protocol for this. Iām not a friend of VOX-controlled amps .
The same I use with my second amp too, a XIEGU XPA125B, too.
73 Heiko, DL1BZ
6 Likes
Thanks for this, Barry. Do you happen to know the code for this size of SMD resistor? I am a surface mount noob - back when I was a lad everything was enormous and soldering was like agriculture
2 Likes
The resistor I used was one from my āJunkā box. I bought a whole lot of SMD resistors from ebay for various projects. I think it was an 0805 size component. 2mm long by 1.2mm wide.
2 Likes
Now 4 activations using this amp and itās working very well after the mods
Iām usually running 2.5W input from the FT-817 and sharing a 3Ah LIPO between the 817 and the amp
It certainly helps catching the S2S on 20M SSB ā¦
Cheers
Rick
4 Likes
Yes thanks everyone in this thread. I had almost given up on this amp assuming there must be some strange common mode current thing or way off output impedance issue. The output impedance does seem to vary from 50ohms on some bands but now itās not tripping out all the time I can poke out above the noise and catch the DX I was missing. 73 Hugh 9V1SA
4 Likes
I got my PA50 a few days ago and made the KD7QOW mod. I tried to use it with my KX3 but it seems that in the auto sensing mode at least, the two donāt play nice together. However, it works reasonably well with my Xiegu X6100. The input power it will tolerate varies significantly, from 1W on 80M to 4W on 17M (CW). I get 40-50W out, depending on the band. Iām still testing.
By the way, there is a translated manual on the Aliexpress website, though the file name is in English. The link below takes you to Dropbox and the file is there. Dropbox - all information - Simplify your life
3 Likes
Hello all. First post! I know this is an aging topic but I am curious if we have a general consensus regarding the best resister replacement or piggyback size / value for R36 to help with the hyper sensitive SWR issue. And if yaāll decided to just replace the resister or run another in parallel with the original. Thatās a two part question. Values please. (all bands are important to me :p)
I just received mine and the board number is v2.5 I will post pictures later today for you guys. Just a long shot, but maybe there is a HW change (wishful thinking)ā¦ I will check FW version when I take the pictures for this thread later tonight.
Thanks in advanceā¦
2 Likes
I have bought one of these recently. I was planning to do the resistor mod, but Iāve not opened it up yet. It has, however, worked straight out of the box with my 20/40m sotabeams linked dipole with no SWR problems. I had a few teething issues getting the unit to auto-switch to TX (which I think was down to me anyway), but so far am impressed with it (it does feel to be good value for money).
73s
Dave.
2 Likes
HW rev: 2.5.6
FW rev: 2.1
Here are the pictures (I took many more if you need them uploaded, let me know):
Standing by for answers to the questions in my prior post.
Thanks in advance!
2 Likes
Bump anyone? I hope my bump doesnāt get me in hot waterā¦
1 Like