Hi, my QMX does not start. When I disconnect all four boards (2x PS, 1x LCD, 1x controls) I measure about 1 Ohm from VDD to ground. Accordingly, a high current is drawn. I applied 1V and at 500 mA I could only see a very small heating at the processor and TXCO with the FLIR. But they did not get hot. I can not see any real hot spot.
?
I then removed L201 and L202. The short circuit remains. I checked all resistors and capacitors connected to VDD such as R211, R216, R218, C217, etc.?
?
All that remains for me is a problem in the area of ICs 501 and 502 or a problem with the multilayer board.
?
Does anyone have similar experiences? I have ordered a new kit and will measure in advance.
?
73, Holger DL9HDA
?
|
Re: Terminal Emulator for Macs
I use screen (/usr/bin/screen) on terminal, like:
$ screen -fn /dev/cu.usbmodem??????
You need to find the device file name of your transceiver. -fn switch is there to disable flow control, since control + Q is used at least in QMX menu.
This way I did not have to install anything new.
|
Re: QDX FDT86256 Mosfet PA Modification
Thanks Kees! I¡¯ll order some immediately to experiment with. Thanks for the heads up.
Tony
Tony,
Mouser has the TN0110 (100V parts). Very nice looking specs relative to the BS170
73 Kees K5BCQ
|
Re: Vacuum-tube final for QRP transmitters?
For the plate supply, consider the $5 boost converter available on ebay (search term "390v dc boost converter").? I used this to provide B+ for a vintage receiver from a LFP battery.? The output voltage is regulated, but it has some LF noise.? If that matters, add a resistor or choke plus another filter capacitor after that.? Also, load it at 5 or 10% of full current between key-up and key-down.? Note: the output is not electrically isolated from the input.
Halden VE7UTS
|
Now on RX bandpass filter.
My design worked largely as intended, although the filter peaks may be a bit off on some bands from 60 to 15m. For some reason, the BPF for 12/10m is behaving very strangely. I used all four channels of the multiplexer for switching two inductors and four capacitors. At leas 60/40m, 30/20m, 17/15m filters seem working fine without strange behavior previously reported.
The receiver sideband rejection, image rejection all look good. I did not change anything in the receiver other than the BPF bank.
|
Me too. We'll see if this has really solved the issue after a week or two of use.
toggle quoted message
Show quoted text
OK Cliff!I'm glad you have it working. I just wish we knew exactly what hadhappened to Q507.JZOn Sat, Sep 23, 2023 at 8:37?PM Cliff <ae5zaham@...> wrote:
Hi John,
Well to pick up the thread we had earlier, I got a new Q507 and installed it.
Tested the modulator for being on based on what Hans said about the fact that before the first transmission the drain shows high, then shuts the modulator off after the first transmission. It does exactly that so apparently the modulator is fine, i.e. not on as we thought, at at least.
Now the gate is 11v before the first transmission then 0.3 after transmission. Maybe the Q507 was bad after all?
Anyway it's working again. Wonder how long the finals will last this time. Hmmm....
Hopefully it's fixed now, but I won't know until I use it for a while.
73,
Cliff, AE5ZA
On Sep 19, 2023, at 14:48, Cliff <ae5zaham@...> wrote:
John,
Thanks for the explanation. I fully understand. No controversy here.
73,
Cliff, AE5ZA
On Sep 19, 2023, at 14:40, John Z <jdzbrozek@...> wrote:
Cliff,
Sorry to hear about the busted FET!
The modulator in QMX serves a number of special functions that QDX does not require.
CW wave shaping, to eliminate annoying ?transmitted key clicks, is one example.
Someday it will be used to create. synthetic SSB.
Hans also uses it to gracefully wind down the transmit power at the end of each transmission. That should avoid a damaging Ldi/dt spike from the PA's power feed inductor.
QDX does not benefit from that spike protection, as it has no modulator circuit. Hence, you will see discussions about the use of commutating diodes, zener diodes, or other means to protect the QDX finals.
These discussions always seem to drag along some controversy. There should be none. The Ldi/dt spike is real and it puts the PA transistors into a dangerous out-of-spec condition.
JZ KJ4A
On Tue, Sep 19, 2023, 2:14 PM Cliff <ae5zaham@...> wrote:
John,
Looking at the schematic for the QDX there is no modulator like the QMX and the L14 which is similar to the L507 in function is always at the input voltage, 11V in this case. At least in the QDX it's normal to have the input voltage on that coil all the time, but I guess not in the QMX.
73,
Cliff, AE5ZA
On Sep 19, 2023, at 11:34, Cliff <ae5zaham@...> wrote:
Thanks John,
Well, the gate had 6.5V on it. The Source and Drain had 11V. Wasn't sure whether that was 6.5 was high or low. Low I was expecting to be 0V. I lifted the Gate pin and now the pad for the gate is still 6 V, but the Drain is 5.4 and the Source is still 11V. It appears as if Q507 was being turned on. Interesting with the gate disconnected from the pad the idle current drain still hasn't dropped. Seems as if that wasn't causing an increase in receive current.
What do you think?
73,
Cliff, AE5ZA
On Sep 19, 2023, at 00:07, John Z <jdzbrozek@...> wrote:
Cliff,
You have confirmed that there is a problem at the PMOSFET modulator.
When all the BS170 are good, they are all off in receive mode and you can expect a normal receive current level even if the modulator is stuck "on". When one or more have failed, a stuck modulator will allow elevated receive current.
Your reading of 11 volts during receive is proof that the modulator is turned on, wrongly.
The next step is to measure the gate voltage at Q507 in receive. If it is low, Q507 is being driven on by the four 'difference amplifier' transistors that compare the DAC signal from the processor to the output of the modulator and produce the gate drive signal for Q507. Determining which have failed would be the next challenge.
If the gate voltage us high, and Q507 is conducting, Q507 is shorted.
...
I see a preceding post that goes on a remarkable run-on over your use of sockets. There is a grain of truth in there, amongst several misconceptions, in that socketing does somewhat impede heat flow out the leads of the BS170. That was neither your initial problem nor your biggest problem now.
In the future I would advise a thin layer of thermal compound under the transistors, and between them and the washer. too. Do not over tighten the nut and screw as thermally expanding transistors will be squeezed badly.
The molded case of modern TO-92 transistors contains a thermally conductive filler to allow the case to participate in heat rejection. The via holes that pepper the plated pad that the transistors sit upon are not helping that at all. They impart surface irregularities ( and hence air gaps) that get in the way of good heat rejection. A little thermal compound helps overcome that.
Good luck, JZ KJ4A
On Mon, Sep 18, 2023, 11:14 PM Cliff <ae5zaham@...> wrote:
Good thoughts JZ.
When I see the idle current go up I open things up and find one of the BS170s is causing it. Pull the bad one and it goes back to normal receive (idle) current so I doubt that the modulator is on, but will verify it anyway.
In receive has 11V on it. Does that mean that Q507 is bad? Sorry I'm not familiar with how Mosfets work? Could the very bad current draw when I had left off the antenna caused that? That looks like a pain to remove without a hot air rework system which I don't have.
73,
Cliff, AE5ZA
On Sep 18, 2023, at 21:05, John Z <jdzbrozek@...> wrote:
Cliff,
I think you are correct when you suggest that something more is going on with the rig.
I suspect that the PMOSFET modulator may be stuck "on". It would not afford protection against destructive Ldi/dt spikes from L502 if it were stuck.
You mention idle current going up when BS170s fail. I presume that you mean current drawn while in receive. That would be another indication of a fault in the modulator section. The modulator should take PA current to zero in receive.
If you see voltage on L502 during receive, that would confirm a stuck modulator.
You might be able to go further in diagnosis using just a multimeter, but at that point it may be better to send your QMX to Jeffrey W. Moore for repair.
JZ KJ4A
On Mon, Sep 18, 2023, 8:46 PM Cliff <ae5zaham@...> wrote:
Hi Paul,
Thanks for the response.
I've had it fail while in different modes. Olivia usually has the longest key down times since it's quite slow. This last time I was in Feld Hell which is one of the easiest modes there are on the rig. It's not a constant transmission, but many with very short pauses mixed in. The rig had been sitting for several hours so it wasn't even hot.
No scope here, sad to say. Testing the Q503 with digital VM and verified 2.5 volts on all final gates.
You raise an interesting point about spikes. Maybe the first time it was damaged was bad enough that it damaged something so there are spikes that are damaging the finals. What that would be I don't know, but maybe one of the experts on this list may have an idea.
Power and SWR are the same on a dummy load. Could try simulating a QSO with the dummy load, but it might take a long time to get enough time on the rig to act up.
The Power/SWR Meter is built into the dummy load and is in between the tuner and the rig so I take it the SWR is what the rig sees. Putting the SWR meter after the tuner shows 1.3:1 SWR. Tuning with the LED indicator in tune mode ends up with the same settings on the tuner and the current draw is the same so I assume power out is the same also.
I agree, something I/we don't understand is going on.
As it is now I'm getting reticent to use the QMX as it's unreliable these days. I have 2 QDX rigs I could use, but prefer the QMX if it can be made reliable by figuring out what is wrong. Of course future firmware upgrades should make it even better.
73,
Cliff, AE5ZA
On Sep 18, 2023, at 18:04, Paul - AI7JR <paul.hanchett@...> wrote:
You mentioned Olivia-- Is that the predominate mode here? Were there other modes where you had a problem? You mention 8-10 minute key-downs, is that typical of your operations?
Are you able to confirm proper waveforms on the gates and drains of the BS170's? No spikes, proper voltages?
Could you simulate a QSO into a dummy load, do the currents look the same? You mention the SWR is 1.0, is the load impedance presented to the QMX actually 50 ohms, or something else? (Could it be that the output impedance of the QMX isn't really 50 ohms --likely, really-- and when we adjust for 1:1 SWR, maybe the antenna isn't being adjusted to be 50 ohms?)
Your experience suggests to me that we don't understand everything about what's going on!
Paul -- AI7JR
On 9/18/23 14:53, Cliff wrote:
I built this QMX within a a month of the release of the kit. It worked great for 2 1/2 months or so. Solid even with long key down modes of 8 to 10 minutes. Power out about 4 watts. ?Then I unknowingly left the antenna disconnected and called CQ 3 or 4 times in Olivia mode. I happened to look at the Buck/Boost Converter and noticed it was in current limiting mode so immediately shut off the power. Damage done. <Sigh .....>
I installed sockets for the finals and replaced all 4 BS170s and the driver IC503. Worked fine for an hour or maybe 2, btw SWR is 1.0 with a manual tuner. Then, with the antenna attached, it again blew a final during a QSO and showed current limiting. Replaced the bad BS170 and all seemed fine. IC503 looked good. This scenario ?has repeated a number of times now. Replace the bad BS170, check IC503 and it'll run fine for an hour or two of QSOs. I even replaced IC503 and all the finals again just to be sure all was ok.
I've also seen twice where the idle current increased from 130 to 200 ma. That was an indicator of a bad BS170 also.
I'm being forced to conclude that there is something else going on besides the finals just dying because of operator error. All parts are from Mouser so quality should be good. Power out is still about 4 - 4.2 watts. Current is about 750 ma, sometimes nearly 800ma (the increase puzzles me) and voltage is 11.0 volts. SWR is 1.0.
No physical shorts that I can see. It's also not always the same BS170 position that goes bad. The time before last it was Q505 and just now it was Q504.
I sure hope someone has an idea of what can be going on. I've been running out of the case to be sure it wasn't heat build up.
Any and all suggestions welcome.
Thanks.
73,
Cliff, AE5ZA
|
Using my filter and transmitting at 18.068MHz, the second harmonic is about -55dBc. This is about 10dB better than the second harmonics of 60m and 30m through the stock filters (they are at about -46dBc) and it is more than 10dB better-than-required harmonic suppression. The price is that the 10m output is a bit lower due to aggressive cutoff frequency. I think the next refinement may be to relax the filter to minimize the attenuation at 10m.
In short, QMX can have one TX LPF for 17m-12m, as mine is working fully, and the same filter can very likely be tweaked to function on 10m fully.
|
So I just got my Mouser package arrived so I assembled QMX for real. I skipped the 80m LPF but installed my filter in place. I also skipped BS170 but installed LDMOS AFM907NT1 x 2 in place. I just uploaded the firmware, looking at the spurious emissions, etc.
First of all, there does not appear to be any parasitic oscillation or instability in using this UHF power transistor in an HF rig without any additional component to tame the excessive gain and super excessive high frequency response. I did take good care in wiring the transistors in dead bug style to minimize unwanted coupling/feedback while maximize the heat dissipation.
I am driving QMX with a 7.5V supply and measuring 34 to 36 dBm on 80/60/40/30/20m (bands that are activated by default).
Second harmonic seems about -27dBc in my setup. (I measured it at 3.5MHz while running through my custom LPF of 30MHz cutoff, so the harmonics are intact.)
I'll figure out how to configure the rig properly and report more later. For those who don't want to bother incremental reports but want to read one story, I intend to write up something once everything is done and tested in the field.
If anyone else is experimenting many-band QMX modifications, I recommend putting the experimental filter in L510/L513 slot rather than what I did. I put my filter in the 80m slot, since I wanted to leave stock filters for 60/40/30/20m and didn't want to shuffle too many things on the board. However, the 80m filter slot (L511/L506) is a lot more crowded than where 20m stock filter is located. I ended up having to solder SMD capacitors sideways on the back of the board to implement my fifth order elliptic filter (needs one more cap than the stock filter topology). I think I never soldered SMD capacitors sideways before.
Also, if anyone is experimenting the receiver BPF using two separate cores, I would use T30-4 or smaller. I used two T37-10 (one of my stock components) and while they could fit, it looks awkward. The filter Q is so low there that the difference in QU (unloaded Q) is not going to matter.
|
Re: QDX FDT86256 Mosfet PA Modification
Tony,
Mouser has the TN0110 (100V parts). Very nice looking specs relative to the BS170
73 Kees K5BCQ
|
OK Cliff!
I'm glad you have it working. I just wish we knew exactly what had
happened to Q507.
JZ
toggle quoted message
Show quoted text
On Sat, Sep 23, 2023 at 8:37?PM Cliff <ae5zaham@...> wrote:
Hi John,
Well to pick up the thread we had earlier, I got a new Q507 and installed it.
Tested the modulator for being on based on what Hans said about the fact that before the first transmission the drain shows high, then shuts the modulator off after the first transmission. It does exactly that so apparently the modulator is fine, i.e. not on as we thought, at at least.
Now the gate is 11v before the first transmission then 0.3 after transmission. Maybe the Q507 was bad after all?
Anyway it's working again. Wonder how long the finals will last this time. Hmmm....
Hopefully it's fixed now, but I won't know until I use it for a while.
73,
Cliff, AE5ZA
On Sep 19, 2023, at 14:48, Cliff <ae5zaham@...> wrote:
John,
Thanks for the explanation. I fully understand. No controversy here.
73,
Cliff, AE5ZA
On Sep 19, 2023, at 14:40, John Z <jdzbrozek@...> wrote:
Cliff,
Sorry to hear about the busted FET!
The modulator in QMX serves a number of special functions that QDX does not require.
CW wave shaping, to eliminate annoying transmitted key clicks, is one example.
Someday it will be used to create. synthetic SSB.
Hans also uses it to gracefully wind down the transmit power at the end of each transmission. That should avoid a damaging Ldi/dt spike from the PA's power feed inductor.
QDX does not benefit from that spike protection, as it has no modulator circuit. Hence, you will see discussions about the use of commutating diodes, zener diodes, or other means to protect the QDX finals.
These discussions always seem to drag along some controversy. There should be none. The Ldi/dt spike is real and it puts the PA transistors into a dangerous out-of-spec condition.
JZ KJ4A
On Tue, Sep 19, 2023, 2:14 PM Cliff <ae5zaham@...> wrote:
John,
Looking at the schematic for the QDX there is no modulator like the QMX and the L14 which is similar to the L507 in function is always at the input voltage, 11V in this case. At least in the QDX it's normal to have the input voltage on that coil all the time, but I guess not in the QMX.
73,
Cliff, AE5ZA
On Sep 19, 2023, at 11:34, Cliff <ae5zaham@...> wrote:
Thanks John,
Well, the gate had 6.5V on it. The Source and Drain had 11V. Wasn't sure whether that was 6.5 was high or low. Low I was expecting to be 0V. I lifted the Gate pin and now the pad for the gate is still 6 V, but the Drain is 5.4 and the Source is still 11V. It appears as if Q507 was being turned on. Interesting with the gate disconnected from the pad the idle current drain still hasn't dropped. Seems as if that wasn't causing an increase in receive current.
What do you think?
73,
Cliff, AE5ZA
On Sep 19, 2023, at 00:07, John Z <jdzbrozek@...> wrote:
Cliff,
You have confirmed that there is a problem at the PMOSFET modulator.
When all the BS170 are good, they are all off in receive mode and you can expect a normal receive current level even if the modulator is stuck "on". When one or more have failed, a stuck modulator will allow elevated receive current.
Your reading of 11 volts during receive is proof that the modulator is turned on, wrongly.
The next step is to measure the gate voltage at Q507 in receive. If it is low, Q507 is being driven on by the four 'difference amplifier' transistors that compare the DAC signal from the processor to the output of the modulator and produce the gate drive signal for Q507. Determining which have failed would be the next challenge.
If the gate voltage us high, and Q507 is conducting, Q507 is shorted.
...
I see a preceding post that goes on a remarkable run-on over your use of sockets. There is a grain of truth in there, amongst several misconceptions, in that socketing does somewhat impede heat flow out the leads of the BS170. That was neither your initial problem nor your biggest problem now.
In the future I would advise a thin layer of thermal compound under the transistors, and between them and the washer. too. Do not over tighten the nut and screw as thermally expanding transistors will be squeezed badly.
The molded case of modern TO-92 transistors contains a thermally conductive filler to allow the case to participate in heat rejection. The via holes that pepper the plated pad that the transistors sit upon are not helping that at all. They impart surface irregularities ( and hence air gaps) that get in the way of good heat rejection. A little thermal compound helps overcome that.
Good luck, JZ KJ4A
On Mon, Sep 18, 2023, 11:14 PM Cliff <ae5zaham@...> wrote:
Good thoughts JZ.
When I see the idle current go up I open things up and find one of the BS170s is causing it. Pull the bad one and it goes back to normal receive (idle) current so I doubt that the modulator is on, but will verify it anyway.
In receive has 11V on it. Does that mean that Q507 is bad? Sorry I'm not familiar with how Mosfets work? Could the very bad current draw when I had left off the antenna caused that? That looks like a pain to remove without a hot air rework system which I don't have.
73,
Cliff, AE5ZA
On Sep 18, 2023, at 21:05, John Z <jdzbrozek@...> wrote:
Cliff,
I think you are correct when you suggest that something more is going on with the rig.
I suspect that the PMOSFET modulator may be stuck "on". It would not afford protection against destructive Ldi/dt spikes from L502 if it were stuck.
You mention idle current going up when BS170s fail. I presume that you mean current drawn while in receive. That would be another indication of a fault in the modulator section. The modulator should take PA current to zero in receive.
If you see voltage on L502 during receive, that would confirm a stuck modulator.
You might be able to go further in diagnosis using just a multimeter, but at that point it may be better to send your QMX to Jeffrey W. Moore for repair.
JZ KJ4A
On Mon, Sep 18, 2023, 8:46 PM Cliff <ae5zaham@...> wrote:
Hi Paul,
Thanks for the response.
I've had it fail while in different modes. Olivia usually has the longest key down times since it's quite slow. This last time I was in Feld Hell which is one of the easiest modes there are on the rig. It's not a constant transmission, but many with very short pauses mixed in. The rig had been sitting for several hours so it wasn't even hot.
No scope here, sad to say. Testing the Q503 with digital VM and verified 2.5 volts on all final gates.
You raise an interesting point about spikes. Maybe the first time it was damaged was bad enough that it damaged something so there are spikes that are damaging the finals. What that would be I don't know, but maybe one of the experts on this list may have an idea.
Power and SWR are the same on a dummy load. Could try simulating a QSO with the dummy load, but it might take a long time to get enough time on the rig to act up.
The Power/SWR Meter is built into the dummy load and is in between the tuner and the rig so I take it the SWR is what the rig sees. Putting the SWR meter after the tuner shows 1.3:1 SWR. Tuning with the LED indicator in tune mode ends up with the same settings on the tuner and the current draw is the same so I assume power out is the same also.
I agree, something I/we don't understand is going on.
As it is now I'm getting reticent to use the QMX as it's unreliable these days. I have 2 QDX rigs I could use, but prefer the QMX if it can be made reliable by figuring out what is wrong. Of course future firmware upgrades should make it even better.
73,
Cliff, AE5ZA
On Sep 18, 2023, at 18:04, Paul - AI7JR <paul.hanchett@...> wrote:
You mentioned Olivia-- Is that the predominate mode here? Were there other modes where you had a problem? You mention 8-10 minute key-downs, is that typical of your operations?
Are you able to confirm proper waveforms on the gates and drains of the BS170's? No spikes, proper voltages?
Could you simulate a QSO into a dummy load, do the currents look the same? You mention the SWR is 1.0, is the load impedance presented to the QMX actually 50 ohms, or something else? (Could it be that the output impedance of the QMX isn't really 50 ohms --likely, really-- and when we adjust for 1:1 SWR, maybe the antenna isn't being adjusted to be 50 ohms?)
Your experience suggests to me that we don't understand everything about what's going on!
Paul -- AI7JR
On 9/18/23 14:53, Cliff wrote:
I built this QMX within a a month of the release of the kit. It worked great for 2 1/2 months or so. Solid even with long key down modes of 8 to 10 minutes. Power out about 4 watts. Then I unknowingly left the antenna disconnected and called CQ 3 or 4 times in Olivia mode. I happened to look at the Buck/Boost Converter and noticed it was in current limiting mode so immediately shut off the power. Damage done. <Sigh .....>
I installed sockets for the finals and replaced all 4 BS170s and the driver IC503. Worked fine for an hour or maybe 2, btw SWR is 1.0 with a manual tuner. Then, with the antenna attached, it again blew a final during a QSO and showed current limiting. Replaced the bad BS170 and all seemed fine. IC503 looked good. This scenario has repeated a number of times now. Replace the bad BS170, check IC503 and it'll run fine for an hour or two of QSOs. I even replaced IC503 and all the finals again just to be sure all was ok.
I've also seen twice where the idle current increased from 130 to 200 ma. That was an indicator of a bad BS170 also.
I'm being forced to conclude that there is something else going on besides the finals just dying because of operator error. All parts are from Mouser so quality should be good. Power out is still about 4 - 4.2 watts. Current is about 750 ma, sometimes nearly 800ma (the increase puzzles me) and voltage is 11.0 volts. SWR is 1.0.
No physical shorts that I can see. It's also not always the same BS170 position that goes bad. The time before last it was Q505 and just now it was Q504.
I sure hope someone has an idea of what can be going on. I've been running out of the case to be sure it wasn't heat build up.
Any and all suggestions welcome.
Thanks.
73,
Cliff, AE5ZA
|
Hi John,
Well to pick up the thread we had earlier, I got a new Q507 and installed it.?
Tested the modulator for being on based on what Hans said about the fact that before the first transmission the drain shows high, then shuts the modulator off after the first transmission. It does exactly that so apparently the modulator is fine, i.e. not on as we thought, at at least.
Now the gate is 11v before the first transmission then 0.3 after transmission. Maybe the Q507 was bad after all?
Anyway it's working again. Wonder how long the finals will last this time. Hmmm....
Hopefully it's fixed now, but I won't know until I use it for a while.
toggle quoted message
Show quoted text
John,
Thanks for the explanation. I fully understand. No controversy here.
Cliff,
Sorry to hear about the busted FET! The modulator in QMX serves a number of special functions that QDX does not require.
CW wave shaping, to eliminate annoying? transmitted key clicks, is one example. Someday it will be used to create. synthetic SSB.
Hans also uses it to gracefully wind down the transmit power at the end of each transmission. That should avoid a damaging Ldi/dt spike from the PA's power feed inductor.
QDX does not benefit from that spike protection, as it has no modulator circuit. Hence, you will see discussions about the use of commutating diodes, zener diodes, or other means to protect the QDX finals.
These discussions always seem to drag along some controversy. There should be none. The Ldi/dt spike is real and it puts the PA transistors into a dangerous out-of-spec condition.
JZ KJ4A? John,
Looking at the schematic for the QDX there is no modulator like the QMX and the L14 which is similar to the L507 in function is always at the input voltage, 11V in this case. At least in the QDX it's normal to have the input voltage on that coil all the time, but I guess not in the QMX.
Thanks John,
Well, the gate had 6.5V on it. The Source and Drain had 11V. Wasn't sure whether that was 6.5 was high or low. Low I was expecting to be 0V. I lifted the Gate pin and now the pad for the gate is still 6 V, but the Drain is 5.4 and the Source is still 11V. It appears as if Q507 was being turned on. Interesting with the gate disconnected from the pad the idle current drain still hasn't dropped. Seems as if that wasn't causing an increase in receive current.
What do you think?
Cliff,
You have confirmed that there is a problem at the PMOSFET modulator.
When all the BS170 are good, they are all off in receive mode and you can expect a normal receive current level even if the modulator is stuck "on". When one or more have failed, a stuck modulator will allow elevated receive current.
Your reading of 11 volts during receive is proof that the modulator is turned on, wrongly.
The next step is to measure the gate voltage at Q507 in receive. If it is low, Q507 is being driven on by the four 'difference amplifier' transistors that compare the DAC signal from the processor to the output of the modulator and produce the gate drive signal for Q507. Determining which have failed would be the next challenge.
?If the gate voltage us high, and Q507 is conducting, Q507 is shorted.
...
I see a preceding post that goes on a remarkable run-on over your use of sockets. There is a grain of truth in there, amongst several misconceptions, in that socketing does somewhat impede heat flow out the leads of the BS170. That was neither your initial problem nor your biggest problem now.?
In the future I would advise a thin layer of thermal compound under the transistors, and between them and the washer. too. Do not over tighten the nut and screw as thermally expanding transistors will be squeezed badly.
The molded case of modern TO-92 transistors contains a thermally conductive filler to allow the case to participate in heat rejection. The via holes that pepper the plated pad that the transistors sit upon are not helping that at all. They impart surface irregularities ( and hence air gaps) that get in the way of good heat rejection. A little thermal compound helps overcome that.
Good luck, JZ KJ4A? Good thoughts JZ.
When I see the idle current go up I open things up and find one of the BS170s is causing it. Pull the bad one and it goes back to normal receive (idle) current so I doubt that the modulator is on, but will verify it anyway.
In receive has 11V on it. Does that mean that Q507 is bad? Sorry I'm not familiar with how Mosfets work? Could the very bad current draw when I had left off the antenna caused that? That looks like a pain to remove without a hot air rework system which I don't have.
Cliff,
I think you are correct when you suggest that something more is going on with the rig.?
I suspect that the PMOSFET modulator may be stuck "on". It would not afford protection against destructive Ldi/dt spikes from L502 if it were stuck.
You mention idle current going up when BS170s fail. I presume that you mean current drawn while in receive. That would be another indication of a fault in the modulator section. The modulator should take PA current to zero in receive.
If you see voltage on L502 during receive, that would confirm a stuck modulator.
You might be able to go further in diagnosis using just a multimeter, but at that point it may be better to send your QMX to Jeffrey W. Moore for repair.
JZ KJ4A? Hi Paul,
Thanks for the response.
I've had it fail while in different modes. Olivia usually has the longest key down times since it's quite slow. This last time I was in Feld Hell which is one of the easiest modes there are on the rig. It's not a constant transmission, but many with very short pauses mixed in. The rig had been sitting for several hours so it wasn't even hot.
No scope here, sad to say. Testing the Q503 with digital VM and verified 2.5 volts on all final gates.
You raise an interesting point about spikes. Maybe the first time it was damaged was bad enough that it damaged something so there are spikes that are damaging the finals. What that would be I don't know, but maybe one of the experts on this list may have an idea.
Power and SWR are the same on a dummy load. Could try simulating a QSO with the dummy load, but it might take a long time to get enough time on the rig to act up.
The Power/SWR Meter is built into the dummy load and is in between the tuner and the rig so I take it the SWR is what the rig sees. Putting the SWR meter after the tuner shows 1.3:1 SWR. Tuning with the LED indicator in tune mode ends up with the same settings on the tuner and the current draw is the same so I assume power out is the same also.
I agree, something I/we don't understand is going on.
As it is now I'm getting reticent to use the QMX as it's unreliable these days. I have 2 QDX rigs I could use, but prefer the QMX if it can be made reliable by figuring out what is wrong. Of course future firmware upgrades should make it even better.
You mentioned Olivia-- Is that the predominate mode here? Were there other modes where you had a problem? You mention 8-10 minute key-downs, is that typical of your operations? Are you able to confirm proper waveforms on the gates and drains of the BS170's? No spikes, proper voltages? Could you simulate a QSO into a dummy load, do the currents look the same? You mention the SWR is 1.0, is the load impedance presented to the QMX actually 50 ohms, or something else? (Could it be that the output impedance of the QMX isn't really 50 ohms --likely, really-- and when we adjust for 1:1 SWR, maybe the antenna isn't being adjusted to be?50?ohms?) Your experience suggests to me that we don't understand everything about what's going on! Paul -- AI7JR
On 9/18/23 14:53, Cliff wrote:
I built this QMX within a a month of the release of the kit. It worked great for 2 1/2 months or so. Solid even with long key down modes of 8 to 10 minutes. Power out about 4 watts. Then I unknowingly left the antenna disconnected and called CQ 3 or 4 times in Olivia mode. I happened to look at the Buck/Boost Converter and noticed it was in current limiting mode so immediately shut off the power. Damage done. <Sigh .....>
I installed sockets for the finals and replaced all 4 BS170s and the driver IC503. Worked fine for an hour or maybe 2, btw SWR is 1.0 with a manual tuner. Then, with the antenna attached, it again blew a final during a QSO and showed current limiting. Replaced the bad BS170 and all seemed fine. IC503 looked good. This scenario has repeated a number of times now. Replace the bad BS170, check IC503 and it'll run fine for an hour or two of QSOs. I even replaced IC503 and all the finals again just to be sure all was ok.
I've also seen twice where the idle current increased from 130 to 200 ma. That was an indicator of a bad BS170 also.
I'm being forced to conclude that there is something else going on besides the finals just dying because of operator error. All parts are from Mouser so quality should be good. Power out is still about 4 - 4.2 watts. Current is about 750 ma, sometimes nearly 800ma (the increase puzzles me) and voltage is 11.0 volts. SWR is 1.0.
No physical shorts that I can see. It's also not always the same BS170 position that goes bad. The time before last it was Q505 and just now it was Q504.
I sure hope someone has an idea of what can be going on. I've been running out of the case to be sure it wasn't heat build up.
Any and all suggestions welcome.
Thanks.
73,
Cliff, AE5ZA
|
Re: QDX FDT86256 Mosfet PA Modification
Exactly Steven. We want a good current burst of 100 to 200 mA at both turn on (sourcing) and turn off (sinking). Resistors in the driver path is not our friend for fast switching. But hey, good try Steve!
Tony - AC9QY
On Sat, Sep 23, 2023 at 5:23 PM Steven Dick, K1RF < sbdick@...> wrote: Steve, Good thought but you will likely have asymmetrical rise and fall times with an active pull-down and passive pullups which gets worse as you go up in frequency.? In this application, symmetric rise and fall times are very important. In addition, the resistors would have to be fairly small, dissipating a lot of power as well as causing additional power dissipation in the driver compared to a driver with an active pull-up and pull-down.
-Steve K1RF??
------ Original Message ------
Date 9/23/2023 5:59:05 PM
Subject Re: [QRPLabs] QDX FDT86256 Mosfet PA Modification
If you run open collector AND gates you could tie the pull up resistor to a larger voltage.? You'd need to wire pull up resistors on all gate outputs though.
Steve KY4GX
Glory to God in the highest, and on earth peace, good will toward men. Luke 2:14
On Saturday, September 23, 2023 at 05:38:36 PM EDT, Tony Scaminaci < tonyscam@...> wrote:
I agree with John regarding the Vgs threshold issues. Microchip¡¯s TN0110 (100V) is the device I found to be most suitable for use as a replacement for the BS170. Unfortunately, last I checked, they were unavailable but there was plenty of stock for the 60V version, TN0160. TN0110 should be back in stock no later than November.
I strongly believe that any of the devices we¡¯ve been discussing, including the BS170, would perform much better being driven from a 7V square wave. Too bad we¡¯re tied to 5V logic for the gate drive. We need a good 100mA drive pulse into each of the gates for good switching times but the ACT08 driver can¡¯t source anywhere near enough current, even with the outputs paralleled. As a result, John showed that we end up with about 4.5V at the gate due to drop in the ACT08¡¯s drivers. This makes things worse because we¡¯re only 1.5V above the worst-case BS170 gate threshold of 3V. The TN devices provide 2.5V of threshold margin at their worst-case which turns them on harder and increases power output.?
I ran a few sims with various bipolar drivers in emitter follower configurations but achieved only marginal improvement. I¡¯m still looking for a driver that can switch at 30MHz and pump out a hundred mils to charge up the gate capacitance at turn on and also discharge it at turn off. Anyone? Buehler?
Tony - AC9QY On Sat, Sep 23, 2023 at 3:02 PM Ryuji Suzuki AB1WX < ab1wx@...> wrote: JZ, that's basically my conclusion when I was considering QCX fitted with FTD86256 a few years ago. Not exactly a show stopper because I was ok with a small annex board, but effort-benefit tradeoff was poor in my view. It is a very good switching transistor, though.
On Sat, Sep 23, 2023 at 01:57 PM, John Z wrote:
While it is a superb
choice for a clean sheet project, it would not be my choice for an
upgrade to QDX or QMX. The Vth spec is a show-stopper.
|
Re: QDX FDT86256 Mosfet PA Modification
Hi Steven, yes very true.? I do that frequently, shoot at the hip only to get shot down in the details!
Steve KY4GX
Glory to God in the highest, and on earth peace, good will toward men. Luke 2:14
On Saturday, September 23, 2023 at 06:23:04 PM EDT, Steven Dick, K1RF <sbdick@...> wrote:
Steve, Good thought but you will likely have asymmetrical rise and fall times with an active pull-down and passive pullups which gets worse as you go up in frequency.? In this application, symmetric rise and fall times are very important. In addition, the resistors would have to be fairly small, dissipating a lot of power as well as causing additional power dissipation in the driver compared to a driver with an active pull-up and pull-down.
-Steve K1RF??
------ Original Message ------
Date 9/23/2023 5:59:05 PM
Subject Re: [QRPLabs] QDX FDT86256 Mosfet PA Modification
If you run open collector AND gates you could tie the pull up resistor to a larger voltage.? You'd need to wire pull up resistors on all gate outputs though.
Steve KY4GX
Glory to God in the highest, and on earth peace, good will toward men. Luke 2:14
On Saturday, September 23, 2023 at 05:38:36 PM EDT, Tony Scaminaci < tonyscam@...> wrote:
I agree with John regarding the Vgs threshold issues. Microchip¡¯s TN0110 (100V) is the device I found to be most suitable for use as a replacement for the BS170. Unfortunately, last I checked, they were unavailable but there was plenty of stock for the 60V version, TN0160. TN0110 should be back in stock no later than November.
I strongly believe that any of the devices we¡¯ve been discussing, including the BS170, would perform much better being driven from a 7V square wave. Too bad we¡¯re tied to 5V logic for the gate drive. We need a good 100mA drive pulse into each of the gates for good switching times but the ACT08 driver can¡¯t source anywhere near enough current, even with the outputs paralleled. As a result, John showed that we end up with about 4.5V at the gate due to drop in the ACT08¡¯s drivers. This makes things worse because we¡¯re only 1.5V above the worst-case BS170 gate threshold of 3V. The TN devices provide 2.5V of threshold margin at their worst-case which turns them on harder and increases power output.?
I ran a few sims with various bipolar drivers in emitter follower configurations but achieved only marginal improvement. I¡¯m still looking for a driver that can switch at 30MHz and pump out a hundred mils to charge up the gate capacitance at turn on and also discharge it at turn off. Anyone? Buehler?
Tony - AC9QY On Sat, Sep 23, 2023 at 3:02 PM Ryuji Suzuki AB1WX < ab1wx@...> wrote: JZ, that's basically my conclusion when I was considering QCX fitted with FTD86256 a few years ago. Not exactly a show stopper because I was ok with a small annex board, but effort-benefit tradeoff was poor in my view. It is a very good switching transistor, though.
On Sat, Sep 23, 2023 at 01:57 PM, John Z wrote:
While it is a superb
choice for a clean sheet project, it would not be my choice for an
upgrade to QDX or QMX. The Vth spec is a show-stopper.
|
Re: QDX FDT86256 Mosfet PA Modification
Steve, Good thought but you will likely have asymmetrical rise and fall times with an active pull-down and passive pullups which gets worse as you go up in frequency.? In this application, symmetric rise and fall times are very important. In addition, the resistors would have to be fairly small, dissipating a lot of power as well as causing additional power dissipation in the driver compared to a driver with an active pull-up and pull-down.
-Steve K1RF??
------ Original Message ------
Date 9/23/2023 5:59:05 PM
Subject Re: [QRPLabs] QDX FDT86256 Mosfet PA Modification
If you run open collector AND gates you could tie the pull up resistor to a larger voltage.? You'd need to wire pull up resistors on all gate outputs though.
Steve KY4GX
Glory to God in the highest, and on earth peace, good will toward men. Luke 2:14
On Saturday, September 23, 2023 at 05:38:36 PM EDT, Tony Scaminaci < tonyscam@...> wrote:
I agree with John regarding the Vgs threshold issues. Microchip¡¯s TN0110 (100V) is the device I found to be most suitable for use as a replacement for the BS170. Unfortunately, last I checked, they were unavailable but there was plenty of stock for the 60V version, TN0160. TN0110 should be back in stock no later than November.
I strongly believe that any of the devices we¡¯ve been discussing, including the BS170, would perform much better being driven from a 7V square wave. Too bad we¡¯re tied to 5V logic for the gate drive. We need a good 100mA drive pulse into each of the gates for good switching times but the ACT08 driver can¡¯t source anywhere near enough current, even with the outputs paralleled. As a result, John showed that we end up with about 4.5V at the gate due to drop in the ACT08¡¯s drivers. This makes things worse because we¡¯re only 1.5V above the worst-case BS170 gate threshold of 3V. The TN devices provide 2.5V of threshold margin at their worst-case which turns them on harder and increases power output.?
I ran a few sims with various bipolar drivers in emitter follower configurations but achieved only marginal improvement. I¡¯m still looking for a driver that can switch at 30MHz and pump out a hundred mils to charge up the gate capacitance at turn on and also discharge it at turn off. Anyone? Buehler?
Tony - AC9QY On Sat, Sep 23, 2023 at 3:02 PM Ryuji Suzuki AB1WX < ab1wx@...> wrote: JZ, that's basically my conclusion when I was considering QCX fitted with FTD86256 a few years ago. Not exactly a show stopper because I was ok with a small annex board, but effort-benefit tradeoff was poor in my view. It is a very good switching transistor, though.
On Sat, Sep 23, 2023 at 01:57 PM, John Z wrote:
While it is a superb
choice for a clean sheet project, it would not be my choice for an
upgrade to QDX or QMX. The Vth spec is a show-stopper.
|
Re: QDX FDT86256 Mosfet PA Modification
There are a lot of fine MOSFET drivers and other ICs that are usable for this, and they cost a dollar or two a piece, but most of them are SO8 or smaller. At that point, why not going LDMOS is one question I wish I had an answer earlier.
|
Re: QDX FDT86256 Mosfet PA Modification
If you run open collector AND gates you could tie the pull up resistor to a larger voltage.? You'd need to wire pull up resistors on all gate outputs though.
Steve KY4GX
Glory to God in the highest, and on earth peace, good will toward men. Luke 2:14
On Saturday, September 23, 2023 at 05:38:36 PM EDT, Tony Scaminaci <tonyscam@...> wrote:
I agree with John regarding the Vgs threshold issues. Microchip¡¯s TN0110 (100V) is the device I found to be most suitable for use as a replacement for the BS170. Unfortunately, last I checked, they were unavailable but there was plenty of stock for the 60V version, TN0160. TN0110 should be back in stock no later than November.
I strongly believe that any of the devices we¡¯ve been discussing, including the BS170, would perform much better being driven from a 7V square wave. Too bad we¡¯re tied to 5V logic for the gate drive. We need a good 100mA drive pulse into each of the gates for good switching times but the ACT08 driver can¡¯t source anywhere near enough current, even with the outputs paralleled. As a result, John showed that we end up with about 4.5V at the gate due to drop in the ACT08¡¯s drivers. This makes things worse because we¡¯re only 1.5V above the worst-case BS170 gate threshold of 3V. The TN devices provide 2.5V of threshold margin at their worst-case which turns them on harder and increases power output.?
I ran a few sims with various bipolar drivers in emitter follower configurations but achieved only marginal improvement. I¡¯m still looking for a driver that can switch at 30MHz and pump out a hundred mils to charge up the gate capacitance at turn on and also discharge it at turn off. Anyone? Buehler?
Tony - AC9QY
On Sat, Sep 23, 2023 at 3:02 PM Ryuji Suzuki AB1WX < ab1wx@...> wrote: JZ, that's basically my conclusion when I was considering QCX fitted with FTD86256 a few years ago. Not exactly a show stopper because I was ok with a small annex board, but effort-benefit tradeoff was poor in my view. It is a very good switching transistor, though.
On Sat, Sep 23, 2023 at 01:57 PM, John Z wrote:
While it is a superb
choice for a clean sheet project, it would not be my choice for an
upgrade to QDX or QMX. The Vth spec is a show-stopper.
|
Re: QDX FDT86256 Mosfet PA Modification
Ones I considered back then, in no particular order (some are prob overkill, some are maybe not fast enough to our taste):
ADP3654
IX4426-IX4427-IX4428
IXD_602
TC4451/TC4452
TC4421/TC4422
TC4426/TC4427/TC4428
|
Re: QDX FDT86256 Mosfet PA Modification
I agree with John regarding the Vgs threshold issues. Microchip¡¯s TN0110 (100V) is the device I found to be most suitable for use as a replacement for the BS170. Unfortunately, last I checked, they were unavailable but there was plenty of stock for the 60V version, TN0160. TN0110 should be back in stock no later than November.
I strongly believe that any of the devices we¡¯ve been discussing, including the BS170, would perform much better being driven from a 7V square wave. Too bad we¡¯re tied to 5V logic for the gate drive. We need a good 100mA drive pulse into each of the gates for good switching times but the ACT08 driver can¡¯t source anywhere near enough current, even with the outputs paralleled. As a result, John showed that we end up with about 4.5V at the gate due to drop in the ACT08¡¯s drivers. This makes things worse because we¡¯re only 1.5V above the worst-case BS170 gate threshold of 3V. The TN devices provide 2.5V of threshold margin at their worst-case which turns them on harder and increases power output.?
I ran a few sims with various bipolar drivers in emitter follower configurations but achieved only marginal improvement. I¡¯m still looking for a driver that can switch at 30MHz and pump out a hundred mils to charge up the gate capacitance at turn on and also discharge it at turn off. Anyone? Buehler?
Tony - AC9QY
On Sat, Sep 23, 2023 at 3:02 PM Ryuji Suzuki AB1WX < ab1wx@...> wrote: JZ, that's basically my conclusion when I was considering QCX fitted with FTD86256 a few years ago. Not exactly a show stopper because I was ok with a small annex board, but effort-benefit tradeoff was poor in my view. It is a very good switching transistor, though.
On Sat, Sep 23, 2023 at 01:57 PM, John Z wrote:
While it is a superb
choice for a clean sheet project, it would not be my choice for an
upgrade to QDX or QMX. The Vth spec is a show-stopper.
|
Ok, I'm baffled as as to where to look next. Suggestions welcome. Here are the preconditions:
QMX Built for 12v. Testing at 12.0v.
1_009a firmware
Boots correctly? ?-? 90ma with backlight on/ 80ma off
Tx Power testing? into 51.5 (measured) dummy load
? ? ? Measured peak to peak voltages with scope Siglent 1202X
? ? ? All bands initially had? consistent Tx power? -but low on 30M and 20M
? ? ? 80M 3.5W
? ? ? 60M 3.25W
? ? ? 40M 3.25W
? ? ? 30M 1.9W
? ? ? 20m? 1.25W??? ? ?
All bands (at least initially) have Rx - I think the initial RF sweeps looked fine (but I did not record them)
?
Actions taken:
? ? ?- squeeze / expand 20m?
? ? ?- reflow solder 20 LFP?
? ? ?- reflow solder on all LPF toroids - verified connections
? ? ?- verified C525 and C522
? ? ?No change
? ? ?- reflowed some additional soldered through hole connections
??
Ran Hardware / RF Sweeps
? ? ? Attenuation on 80, 60 and 40 - all peak high outside band - significant in band attenuation
80M
60M
40M
30M? ??
20M
Action taken:
? ? ? ?- Rebuilt? L401 as per Manual rev i
? ? ? -? Reflow solder on T402 and caps surrounding mux chip
? ? ? - measured 2.5v on C406
? ? ? No change to RF sweeps
?
? ? ? - reflowed solder on most through hole connections
Resumed Tx testing
? ? ? - Erratic peak to peak? voltage measurement cycling through band selections,?
? ? ? ? ? ? other than the behavior on 20M, power does not change except after selecting another band
? ? ? ? ? ? ?and returning to the band indicated.?
? ? ? ? ? ? ? ?80M? ?50v? ? or? ? ?37.5v? ? ??6.25W / 3.51W
? ? ? ? ? ? ? ?60M? ?60v? ? or? ? ?35.6v? ? ? 9.0W? /? 9.0W? ?yikes!!!
? ? ? ? ? ? ? ?40M? ?60v? ? or? ? ?59v? ? ? ? ?9.0W /? ?8.7W? ?yikes!!!
? ? ? ? ? ? ? ?30M? ?35v? ? ?steady? ? ? ? ? ? 3.0W? ??
? ? ? ? ? ? ? ?20M? ?62v instantaneously - hits current limiting on supply,? drops to 36v? ?
? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?9.61W? / 3.24W? ?yikes!!!? ? ??
Assumptions
? ? ? - Need to resolve erratic Tx voltage first
Action taken:
? ? ? - Verified Dummy load - 51.5 Ohms, Ran 30M QCX-mini Tx into dummy load - steady at 3.75W at 12v
So, I've successfully taken the QMX from very functional albeit weak 30/20M?Tx performance - to erratic Tx behavior and made it deaf on 3 bands.?
I'd congratulate myself, but this is not the desired outcome!
Help?
Reasonable suggestions welcomed!
? ? ? -?
|
Re: QCX Mini Surface Mount Resistor Disaster
Robert,
HOLY CRAP!?? I just looked at the picture on your QRZ page.? You know it is possible to take this do it yourself thing too far! ;)
--
Chris / W2BPL
|
Ryuji Suzuki AB1WX
Chris / W2BPL