Loss of transmit power

Post Reply
W8TEE
 

Posts: 8
Joined: 06 Mar 2023, 22:03
Location: Cincinnati, OH, USA
Contact:

Loss of transmit power

Post by W8TEE »

All:
I made several contacts with my rig, mostly on CW, and things appeared to go well. I came down several days later, fired up the (tr)uSDX and the receiver was working great. Eventually, I started calling CQ and, after about 10 seconds, the transmitter just quit and power dropped to zero. I was working 40M using an EFHW with SWR showing 1.02:1. Power out was never above 1.9W on any band, which I was willing to live with. I'm powering it from my shack's main JetStream PS (13.8V up to 30A). I tried 20M with a 1.29:1 SWR with the same results. The receiver works consistently well.

I'm really a software guy, so things like this are difficult for me to fix, although I have some fairly good test equipment (signal generator, dual 200mHz scope, DVM, etc). I've opened her up looking for bad solder joints, etc., and don't see anything on the board. I did notice that caps C10, C13, C18, C26, C34, and C39 are missing, which seems like a lot to me, but then I'm nowhere near an EE understanding of things.

If anyone has suggestions, I'm all ears!

Jack, W8TEE
dl6sez
 

Posts: 399
Joined: 30 Dec 2021, 22:54

Re: Loss of transmit power

Post by dl6sez »

Hello OM Jack,

the closest possible thing for me are the PA-FETs 3x BS170. The missing Cs are correct. It sounds like your lpf toroids are without the normally required "tuning of turns", resulting in a too low output power. This could also be mismatch for the FETs at PA. If the harmonic notch is too low and is not good for the FETs than. At 13,8V only 1,9W is too low. Should be at least 2,5W to 3W on some bands. But as i said, mismatch of the class-E PA-Fets can kill them although your swr meter shows a good antenna match because it is an internal bad match of the output of the FETs. So it is not a hard task to remove the 3 FETs. You can check before if they are the reason:
Remove one wire of the single turn winding of T1 from its solder point. Than you can measure with an ohm-meter on the outermost pins on one BS170 (it's Drain and Source, Gate is the middle pin) in both directions. There should be no conduction in one direction. There is also a diode in the FETs included therefore one direction let's the current from your ohm-meter flow. All 3 FETs are paralled so you measure all at the same time. If you found an error you can than remove them easily by cutting the FET feet from the pcb. Afterwards remove the feet pins on the pcb one by one with a solder iron and a tweezer.
You should only replace the BS170s from a well known source as Mouser, Farnell etc..... BS170 are sold by many chinese sellers and other ignorant sellers and are not really good real Onsemi (ex. Fairchild) BS170s.
Normally they are not used in RF PAs ;) they are simple low cost switches for relays and motors....
for this applications everything is functioning with any FET doing like an N-channel enhancement low power device :)

73 de Chris
Chris DL6SEZ, JN48XL near Ulm, Southern Germany
W8TEE
 

Posts: 8
Joined: 06 Mar 2023, 22:03
Location: Cincinnati, OH, USA
Contact:

Re: Loss of transmit power

Post by W8TEE »

Chris:

This is the perfect kind of answer for me: a good place to start with the steps to follow clearly stated. I lack the confidence to do such things on my own...but I'm learning. I sincerely appreciate the time you took to help me. Thanks!

72,
Jack, W8TEE
Ohwenzelph
 

Posts: 207
Joined: 01 Jan 2022, 03:47

Re: Loss of transmit power

Post by Ohwenzelph »

Don’t know if it will help but check out the thread
viewtopic.php?t=480
For most things if they are from the same batch they are likely close enough, but if you want to optimize…
Some of the recommendations come from back in softrock history
Post Reply

Users browsing this forum: No registered users and 2 guests