Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
Search
QMX (not-plus) power issue
|
A couple of weeks ago, I think, I brought up an issue I'd seen with my QMX low-band (20-80), where I was only getting about 1.2W out on 20 meters. Today, I decided to run some tests using a new power meter and dummy load, and also to see if maybe I'd set a setting wrong. The results were a little surprising, and probably point to some hardware fault, and not a setting problem. To double-check that theory, I did a factory reset and a firmware upgrade, and received the same results. I have both narrated video of the power meter, and screen recordings of the terminal app, that I can edit together if they would be useful. I had been thinking of terms of posting a YouTube video about "fixing it" but then, of course, I was not able to fix it! (That said, I may at some point do a YouTube video guided tour of the Terminal App!) What I'm seeing when I hit the 'T' key in the Diagnostics screen: Input voltage looks steady at 11.7 SMPS 3.3V shows 3.3V, 5V shows 5V, Bias SMPS shows 30mA as does PIN Bias For all of these -- and this may be the hint -- the transmitter voltage shows 10.2 transmitting, 0.6 idle 80m: Diagnostics show 4.7W, power meter shows 2.0 60m: Diagnostics show 5W, power meter shows 3 40m: Diagnostics show 4.3W, power meter shows 2.5 30m: Diagnostics show 2.5W, power meter shows 1.25 20m: Diagnostics show 1.2-1.3W, power meter shows 0.5 Where is the most likely place to look for both the shortfall relative to the meter, and the shortfall relative to general expectations of 5W as opposed to 1.2? Thanks! Michael WD0OM |
|
By the way, I will accept the possibility that the meter might need calibration! It's literally brand new, out of the box, and it's a cross-needle type with what look to be calibration screws under the needle display. If we think we can trust the diagnostic's idea of what the power output is (even if it's not what it should be on a given band), I will use that as my standard to calibrate and eliminate one mystery! On Thu, Sep 19, 2024, at 21:19, Michael - WD0OM wrote:
|
|
Some helpful numbers to post would be the voltage you built your QMX for, and what is the overall input current draw in receive mode of your QMX??
?
For a 12v QMX you should see around 90 mA or so of current when it's on and idle (not transmitting).?
?
If it's substantially more than 90mA then your power meter is likely correct, and one of the output power transistors is likely blown/defective.... As a first diagnostic guess.?
--
73 de Roy - KI0ER Littleton, Colorado USA |
|
Just some quick thoughts on your observations.
?
RE: low watts out on 30/20.
They share the LPF pair.
Inspect solder joints, turns on the toroids, capacitors etc.
Fiddle with spreading and squeezing turns on those two as well.
?
RE: power output differences
Have you tried different interconnect cables or BNC couplers?
Do you get the same results on the diagnostics with just the dummy load connected to the QMX? Try with both interconnect cables and/or BNC couplers.
?
Any chance you have a nearby Ham Buddy that can check the new power meter. Or perhaps you have another radio you know works well?
?
Hope it's something simple since the new power meter seems to be showing just about 1/2 of the expected power out?
? |
|
Roy: I did in fact post that it was a 12V QMX. I did not build it, however; I bought it from someone else who had. I have to admit I had forgotten to pay attention to what the power source (an Anker power bank + 12V trigger cable) said my current draw was. That had come up before, so that's an oversight I should correct next time I get a chance. I do suspect it's higher than it should be, though. While I was using the power bank to also run the laptop (long story), that should not have drained this particular power bank to zero in the time I was using them; but it did. Still, I should check the numbers. Not sure when I'll get a chance to do that experiment next, but I'll report back when I do. On Fri, Sep 20, 2024, at 04:55, Roy - KI0ER wrote:
|
|
Yeah, I realized that after I'd posted, then felt dumb about it. I did take a moment to zero them out. It did not really change the output result, of course. On Fri, Sep 20, 2024, at 06:46, Donald S Brant Jr wrote:
|
|
I did actually take the QMX apart after I posted, as much out of curiosity as anything else -- I am building a QMX+, but had not actually looked inside the QMX since I bought it used, already built. It was instructive in general. I didn't see anything obviously bad about the solder joints, but I didn't look too closely at specific components like the 20/30 LPF.? I did not try different cables or connectors for the simple reason that I don't have any. I have exactly one short jumper right now, about 3', and it's brand new. It's UHF male on both sides, so I'm using an elbow coupler to the radio, of which I also have exactly one (coupler, that is, not radio, qv) at this time. One has not had so much time in this hobby as to have acquired lots of spares, yet. I do see the same result on the diagnostics connecting the dummy load directly to the coupler, though. I do have other radios whose performance is known and trustworthy that I can test -- I just didn't have time to fetch them to my makeshift work space last night (I also have neither a shack, nor a proper workshop), but testing the power meter against one of them is on my list. I also do have local people who can help me out if need be. Michael WD0OM On Fri, Sep 20, 2024, at 08:17, Greg McCain via groups.io wrote:
|
|
So, I put it on a buck-boost that could give me a better readout than the power bank on amperage draw. At idle, it's drawing 0.125A; transmitting on 20m, it's drawing 1.25A. At this scale, I'm assuming "0.125A" count as "substantially more" than "0.090A", so I'm betting something's blown! /m On Fri, Sep 20, 2024, at 04:55, Roy - KI0ER wrote:
|
|
Hi Michael.?
?
When I had a blown PA output transistor, my 12v QMX was drawing a little over 200mA on receive/idle mode. So yours is behaving a little differently at 125mA receive current draw.?
?
In my case one of the 4 PA transistors had an internal short from gate to drain. With the display board removed and the QMX powered up, I could feel (and actually see, since I had an infrared camera) one of my 4 PA transistors was getting quite hot during receive, which should not occur.?
?
You may need help from one of the more knowledgeable electronic gurus, like Ludwig, in diagnosing what's going on.?
?
But, if you can detect one of your 4 PA transistors getting warm in receive mode, then it's probably gone bad and needs to be replaced.?
?
This is more of a hit it with a hammer type of diagnosis. You may need more expert help than I can provide.?
--
73 de Roy - KI0ER Littleton, Colorado USA |
|
Roy, thank you. That fits, so far, with what people seem to describe as part of the cluster of issues around the BS170s. Honestly, even if the BS170s turn out not to be the problem, starting with them seems prudent. They appear to be an Achilles heel. They might have come to me with one already blown, or I might well have blown one myself mishandling power to the device either at input (I did experiment with an "ordinary" radio battery just to see how the protection behaved) or at the output (I have used the device with an autotuner, and that's come up as a possible killer of BS170s). I have an infrared thermometer, still in its packaging :-D I bought it specifically for future diagnostics like this one. Probably not quite as good as an infrared camera for this, but possibly better than touch alone. I assume I should remove the retaining washer, or is result something I should be able to easily read with the washer in place? It may be a few days before I can experiment with it again, but absent other advice, I believe this will be my next investigation. Thank you! Michael WD0OM On Sat, Sep 21, 2024, at 05:10, Roy - KI0ER wrote:
|
|
I wouldn't assume you have a problem in the BS170s.
My newly built QMX mini draws 125mA on idle/receive, and all diagnostic tests work correctly, and it seems to receive and transmit correctly.? There may be something wrong that I am not aware of, but all seems ok so far. On transmit it draws about 1.25A.
I also have a cross-needle power/swr meter (MFJ-880), and its power readings match almost exactly the ones from the QMX diagnostic menu.? So I don't know what the issue is there for you.? The QMX reads the power directly from its SWR bridge at the output, so there isn't much in the pathway between that and an external meter except cables and connectors.
It is a well-known fact that the QMX (and QDX and QCX) output lower power on the upper bands.? Full 5W power on the low two bands, and gradually decreasing to 3W or so the top band.? Your 20M power is lower than it should be, but since the other bands seem correct, it is likely not a BS170 issue.? Search the prior messages here for 20M low power or low power and you will find several threads discussing this and what people have done to try and remedy.? I think mostly it is adjusting the position of windings on toroids.? You may also have a bad/wrong value capacitor in the 20M filter which could contribute to this.? Does the LPF sweep and Bandpass filter sweep function from the diagnostic menu for 20M give results that look very much like the ones in the operations manual? If not, the filters are the first place to check. |
|
p.s. to my prior post: oops, my QMX is built for 9V, so 125mA at 9V is about the same as 90mV at 12V from the prior posts.? So you do have a higher than normal idle current draw.
But since you are getting a full 5+W on the lower two bands, I don't know how it can be a problem with the BS170s.? Based on other posts I have read here, I don't think you can expect a full 5W on 20M, either.? There is definitely an issue on 20M and probably 30M on your device - so since those two bands share the same filter, that's the first place I would look.
Stan |
|
I agree with Stan's assessment. When I had a blown PA transistor in my QMX, the rig (12v) was drawing over 200mA during receive.?
?
I didn't have to remove the Heat-sink washer to find the bad one. It was quite obvious which one was bad, but it likely helps to remove it to see if any are warmer than others. Start with a cold radio, and don't transmit at all.?
?
Stating the obvious, you have an extra 35mA going somewhere. Is that constant, or does receive current draw change by band??
?
Hopefully someone else can jump in and help you with diagnostics.?
--
73 de Roy - KI0ER Littleton, Colorado USA |
|
It seems to be constant, regardless of band. I haven't had time to do much other investigation, but I do remember that was the case. /m On Sun, Sep 22, 2024, at 02:41, Roy - KI0ER wrote:
|
|
Based on your power meter readings, if it were me based on my prior experience, I would suspect that one of the PA transistors has died.? One of mine died with a direct internal short which caused the excessive power draw during receive mode. Yours doesn't seem to be pulling as much power during receive (unless your power source readout is a little off) but it could have died a different way internally.
?
I would, if it were mine, remove the display board, remove the heatsink washer, power it up (still attach a dummy load just in case), and feel if one of the 4 transistors is getting warm or hot. It shouldn't take long for a bad one to generate heat if shorted internally.?
?
And if one is getting warmer than others, again in receive mode only, replace it. Then see if you are back in business with about double the power out that you've measured.
?
But if no excessive heat is detected, I'd then remove them and test each one individually.
?
I can't say why the diagnostics screen is telling a different and inconsistent story compared to your power meter. But if there is a hardware failure, then without being an expert myself I could guess that the diagnostics could be misreporting due to that failure.
?
Stating the obvious ... you need to get that baby fixed and on the air! Fingers crossed.
?
--
73 de Roy - KI0ER Littleton, Colorado USA |
Donald S Brant Jr
Joshua KJ7LVZ
to navigate to use esc to dismiss