OK, so after fitting a new Q103 (damaged in the process of finding the short circuit problem..."Short circuit in buck converter QMX") I'm ready to try again.

PSU set to 7.5v current limited to 250mA

Push the left hand encoder, the display lights up and displays a row of squares on the top row.

Two problems:

PCB1

Vcc is 7.5v, inspite of anything else D108 has not clamped Vcc to 5.6v. PWM_5V sits at 5v, no waveform.

LIN_REG_EN signal from the processor is 7.5v (expecting 3.3v?) under these conditions The Q102/Q101 continues to supply the 78M33, the output from the 78M33 is around 3.0v, and feels warm.

PCB2

PWM_3V3 sits at 0v no waveform, Vdd sits at 2.97v, this is no doubt coming from the 78M33.

So what do you think is going on? Has the processor been damaged by the Vcc of 7.5v (current limited to 250mA)? although it is valiantly generating a LIN_REG_EN of 7.5v

No obvious shorts measured or observed

What a disaster!?

Steve G4EDG


?

开云体育

If a 3.3 V processor pin is at 7.5 V, then you are done. You need a new processor, but since that’s not a thing, you need a new kit, and this one will be a good source of spare parts.

Willie

Sadly I think you're right! Imagine the smoke if I'd used a 12v supply and no current limit.

I'd still like to establish how the Vcc supply failed and supplied essentially Vin.

I'm sure I won't be the last to experience this.

Steve

On Sat, Aug 12, 2023 at 05:29 AM, g4edg wrote:

Steve,

PCB2

So what do you think is going on? Has the processor been damaged by the Vcc of 7.5v (current limited to 250mA)? although it is valiantly generating a LIN_REG_EN of 7.5v

Yes,??LIN_REG_EN of 7.5v (in this case V-IN) and VCC of 7.5V is not repairable by the average builder as Willie pointed out. I would also suggest a new QMX kit and not spending more time on this one.

73 Kees K5BCQ?

I posted some checks that can be done on the PSU boards independently of the main board /g/QRPLabs/message/106787
They may help to find out what is going on.

Chris. G5CTH

That's great Chris, thank you.?

Steve

OK, so I've just worked up the enthusiasm to have a second look at this potentially "bricked" QMX

Regarding the signal LIN_REG_EN, am I right in assuming this signal is 0v (a logic 0 from the processor, as opposed to a logic 1 of 3.3v) ??

Voltage at the R101/Q101 gate is +12v (a current limited 7.5v in my case) this is connected directly to pin 88 of the processor, admittedly via a 100k resistor....is it OK to do this to a 3.3v device? irrespective of how briefly.

I'm not getting the LIN_REG_EN signal to hand over the 3.3v supply responsibility from linear to PWM mode...the 78M33 just sits there and gets warm....output of the 78M33 is 3.3v ......VDD is 2.97v. Is 2.97v enough to run the processor? Data sheet says 1.7v.

The second part of the problem is that VCC is at supply voltage....and not clamped by the 5.6v zener.......VCC gets nowhere near the processor, so it cant be that that has damaged (?) or is preventing the processor from functioning.

Any thoughts?

Steve G4EDG

开云体育

OK, so I've just worked up the enthusiasm to have a second look at this potentially "bricked" QMX

Regarding the signal LIN_REG_EN, am I right in assuming this signal is 0v (a logic 0 from the processor, as opposed to a logic 1 of 3.3v) ??

Before the processor is active R101 will pull LIN_REG_EN high which turns Q102 on and this will enable the linear regulator. Q111 is off to stop the SMPS getting involved and Q110 is on to provide a current sink for the SMPS. The MCU will keep it high while it gets the SMPS going and once it is it will pull it low which will turn the linear regulator(Q102) off, the SMPS supply (Q111) on and the SMPS current sink (Q110) off.

Voltage at the R101/Q101 gate is +12v (a current limited 7.5v in my case) this is connected directly to pin 88 of the processor, admittedly via a 100k resistor....is it OK to do this to a 3.3v device? irrespective of how briefly.

Hans will have though of this, check the processor data sheet but I think it will be fine.

I'm not getting the LIN_REG_EN signal to hand over the 3.3v supply responsibility from linear to PWM mode...the 78M33 just sits there and gets warm....output of the 78M33 is 3.3v ......VDD is 2.97v. Is 2.97v enough to run the processor? Data sheet says 1.7v.

Is 2.97V greater than 1.7V?? Yes, it should be enough to run the MCU, if it isn't broken.

The second part of the problem is that VCC is at supply voltage....and not clamped by the 5.6v zener.......VCC gets nowhere near the processor, so it cant be that that has damaged (?) or is preventing the processor from functioning.

This could mean that Q108 has failed short circuit and D108 has failed open circuit. Or something else wrong in the Q106, Q112, Q108 area which is turning it on. The Zener is probably gone in any case.

Any thoughts?

The linear regulator seems to be generating enough voltage to run the MCU but it appears not to be coming to life and turning the SMPS on, however there could be multiple failures in the power supply boards.

It might be worth trying to check the main board independently by removing PCBs 1 and 2, the display board and the control board, connecting USB and applying 3.3V to VDD.? If the MCU is working it should come up as some device on the USB and you may be able to use PuTTY to check more.

Chris, G5CTH

Steve G4EDG

Chris,

I would still like to have Hans verify the 3.3V Linear Regulator to 3.3V SMPS handoff is implemented in the current Firmware. ..............especially with the higher current measurements seen by several people when in Receive mode.

Steve,

Sorry, but with those voltage measurements and those you gave earlier, I'm afraid your board is toast.??

73 Kees K5BCQ

开云体育

Hi Chris and others

I worked my way through all the steps in your guide to prove PCB 1&2 are functioning correctly...both boards passed.

In spite of everyone's conclusion that my processor was toast....I powered the QMX with 5.6v(it wouldn't switch on at 5.0v) and heard the familiar sound of a memory stick being inserted from my PC, and indeed the QMX appeared as a drive in the file manager. I loaded the latest firmware and could see the file was there along with the EPROM file...I'm guessing this indicates the processor is working.

Unfortunately this state did not last long! After 15 minutes or so the QMX was first not recognised as a proper USB device then later not recognised at all.

The processor produces no PWM_3V3 or PWM_5V signals, but sustains the LIN_REG_EN after the encoder button is pressed. It seems to be running from my external current limited supply which is set to 5.5v, intermittently at best.

The display just shows a row of dark boxes on top row.

In my experience a processor either works or it doesn't, so why the patchy behaviour?

Steve G4EDG

开云体育

Hi Chris and others

I worked my way through all the steps in your guide to prove PCB 1&2 are functioning correctly...both boards passed.

I thought that you said that VCC was at the power supply voltage, in my tests if it is on when PWM_5V is low that's a failure.? PWM_5V must turn Q108 on and off as it goes high and low.? If it doesn't the SMPS can't work.

In spite of everyone's conclusion that my processor was toast....I powered the QMX with 5.6v(it wouldn't switch on at 5.0v) and heard the familiar sound of a memory stick being inserted from my PC, and indeed the QMX appeared as a drive in the file manager. I loaded the latest firmware and could see the file was there along with the EPROM file...I'm guessing this indicates the processor is working.

Yes, good news.? AIUI before any firmware is installed the boot software runs off the linear PSU with LIN_REG_EN high.? Once the firmware is installed the switching regulators are enabled and must work.? Hans' firmware may be checking that the switch mode supply

Unfortunately this state did not last long! After 15 minutes or so the QMX was first not recognised as a proper USB device then later not recognised at all.

The processor produces no PWM_3V3 or PWM_5V signals, but sustains the LIN_REG_EN after the encoder button is pressed. It seems to be running from my external current limited supply which is set to 5.5v, intermittently at best.

If the switch mode supplies aren't working properly then it's possible that the processor can see this and not switch to it.

Do you find that your PC can find a serial port and maybe a sound card?? If there's a serial port then Putty may be able to provide information.

The display just shows a row of dark boxes on top row.

The display runs off the 5V VCC supply, if that isn't working properly then the display won't work properly.

In my experience a processor either works or it doesn't, so why the patchy behaviour?

That's above my pay grade.

Chris, G5CTH

Steve G4EDG

I'm going to put another QMX together in the morning after I've had breakfast and some coffee. Should be great fun.

73 Kees K5BCQ

As a final test I fed the QMX with two external regulated supplies, 3.3v and 5.0v and provided for ADC_3V3 and ADC5V with two potential dividers.

Unfortunately, the QMX reacts in exactly the same way as before when using power via PCB 1 & 2.

Top row of display.... dark boxes?

Nothing happens when connecting the QMX to my laptop...all combinations of powering up and connecting the USB cable.

Maybe the firmware dosent like this arrangement....otherwise I pronounce this QMX dead.

73 Steve G4EDG