Re: Fried my qcx+ finals...


Hello Cher VX Yannig,

QCX (and I think in common with many QRP rigs) the "finals" are very sensitive to high SWR and blow very quickly. If you search some recent posts where i had the same problem you will find much advice. in my own experience in blowing QCX finals many times, it is most likely, hopefully, of only any of Q1-Q3 and/or Q6 to be blown. if you have spare of these the easiest and quickest is probably to replace them and see if it them works, and you can try by replacing one by one, as i did not find a way to test in circuit, but i will post some advice given to me which i followed incorrectly due to my lack of experience and i then actually blew up the entire QCX. So please only if you know what it means below, or if like me you had forgotten how transistors work, some videos to look at to see which is gate etc, will be helpful and how to measure the voltage -- not how i did blowing everything up.

As the blowing of QCX finals is a very frequent thing (at least for me and some others here) i am wondering if there wouldn't be some additional modifications that could be done by us to make some kind of protection? We must remember the QCX is an amazing value rig for all which it contains and Hans the inventor has made incredible job to keep the cost so low and not much profit from each rig, so that we benefit the low cost resembling a rig easily 10 times the price. But these cost savings surely unavoidably come without some things such as: RF gain, filter bandwidth adjustability, CW offset adjustment, high SWR protection, Automatic Gain Control. Yet the experienced electronicians in here know how to do some of those things, so I would like to see what can be done about that. 

is there some SWR protection circuit that we can build to include in our QCX? Or if that is not practical is there some circuit we can put in that e.g. has a LED that will light, or not light, if there is mismatch?  Personally i do keep a SWR meter in line (when i did not i was unlucky) but this also is not a full protection: there are times we do not look at it while transmitting, and an intermittent or slight fault may not show up so well, also we could have as you did WSPR and then we are not present. i really think i should find some way for such protection and it would be good if the experts in here could put heads together to which ways this could be done in an easy and fool proof manner for those of us who build QCX only because of the amazing step by step manual of its construction.

i am not saying you are one of those like me Yannig you may understand these things much better, for myself i passed only for love of CW and memorised most of the electronics, in short term memory at a young age, that RAM was long emptied of most of those electronic things. 

i hope you can fix easily, also there is checking on ic3 voltages after you repair, if you are careful, for me i destroyed also pin 9 which however may not have been important to the operation as it was for BPF set up in the menu, once set up, not needed that menu to be working correctly again. 

Here was one helpful advice when i had same problem as you, but i did not follow it correctly and damaged the entire rig, so i should just have replaced the transistors as i had some spare ones:


Lou, if Q6 is blown Open it won't be able to provide DC to Q1,2&3 and will result in no power out. If Q6 is blown Shorted it will provide DC to Q1,2&3 all the time. This may or may not cause problems with the finals as they don't go into conduction until the gates go high so it's possible to get output even in that case. You can check, without transmitting, for voltage on the drains of the finals. If there is 12 volts there on Receive then Q6 is probably shorted, or something in the keying circuits is making it stay ON. If there is no voltage there on receive, a very short key up may show some voltage increase on the collector of Q6 which indicates it might be good.

But, the three finals are in parallel and there is no way to test them individually without pulling them off the board. You can remove one and test it and if it's good it's possible (not 100% likely) that the other two are also good. If it's bad then it's likely that all three are bad.

The easiest way I've found to test components out of the circuit is with a component tester. I'm not recommending any one in particular. Here is one that's available 

73 ... Ron




On 19 Jul 2021, at 19:14, Arv Evans <arvid.evans@...> wrote:

You can test transistors, and MOSFETS, in several different ways.  
For NPN DC gain you only need to connect the collector through 
470 ohms to Vcc, the base through 4.7k and alternatively through 
470K, and the emitter to ground.  This lets you get an approximation 
of DC gain at the operating voltage and calculated base current.

You can also do an in-circuit test to see if the BE voltage is around 
0.6 volts.

Out-of-circuit tests can use a VOM to see if both B-E and C-B show 
diode action of the proper polarity.  Be sure that your VOM does 
not cause over voltage conditions for either of these junctions.

You can build a free-standing circuit that emulates one stage of the 
circuit it will be operating in inside the rig, and scope that circuit for 
proper operation with similar-to-real-world inputs.

Hang in there.  You are coming along in fine shape.  Soon you will 
be one of the "old timers" who are giving advice to other builders of 
this same equipment.



Good luck and let us know if you need more help there are many here able to assist in the steps. 

73 es 77 de Lou

On 24 Jul 2021, at 15:30, Yannig - F4IUJ via <yannig.robert@...> wrote:

Hello all,

I'be managed to fry the PA of my qcx (I think) by unplugging the antenna while the tx was in WSPR mode. 

I want to determine with certainty which components to replace and I am not sure that I understand fully the troubleshooting guide...

So I've determined that the TX now outputs about 200mW in a 50ohms load. What should I do next? 

In advance thank you.

Yannig F4IUJ

Love 30m and CW?

Join to automatically receive all group messages.