| john65b |
I again violated one of the most important rules in DIY - Never Solder while tired...
I had been happily using a pair of Zappulse 2.1 modules and got a a bug in my butt to upgrade the power supply from single supply to dual mono...somehow I tripped up the rail wiring and saw a nice puff of smoke...took out my tweeters of my KEFs too.
Anyway, I unhooked everything and went to sleep...next day I checked out the modules and saw the two resistors in front of the 2.2uf caps by the wiring inputs that looked charred...but all else from first glance looked OK...I reconnected everything (correctly) and got full rail voltage on the outputs.
After emailing LC Audio about some info and the resistor values (they were very protective of telling me anything value-wise) I replaced the two 20k resistors that I saw looked charred...LCAudio tech said he thinks its 20k, but not sure.
Anyway, it did not make a difference since I continue to get full rail voltage on the outputs...All the beefy resistors on the edges of the board look OK and seem to be fine, but wonder what else I could check/replace to get these two modules working again??
I gave up on the LC Audio tech, as all they could do is credit me $30 each module on a trade in for the newer modules...I didn't really like that option, and I have since built a UCD400AD that I am very happy with, but would still like to get these Zappulses working again...
Any help? |
|
|
| zkaiser |
| How about nice big, high res. picture of demaged module ? Top and bottop ofcourse... |
|
|
| john65b |
No problem, I will post a pic later today...
zkaiser --> are you familiar with this type of issue?? |
|
|
| zkaiser |
| Actually I was doing some rev. eng. on these modules. |
|
|
| john65b |
Perfect!
I liked the LCAudio, but I really like the UCD400 better.
Just would like to get these working again...
Ya know, I might be able to send them to you so you can "do your thing" if you think you could advise me on how to get them operational again... |
|
|
| john65b |
A few pics (crappy, but the wife has the good camera)
You can see the single resistor that was in front of the cap to the left... |
|
|
| john65b |
| And the last one...anyone care to point out what the likely culprits are? The connections were kinda crappy when I got it and left t as is (bought from fellow DIYer here on this thread...) |
|
|
| BWRX |
| Are the FETs damaged? Hint: check for a short circuit. |
|
|
| john65b |
| The FETs are not shorted. |
|
|
| BWRX |
| Well, you're lucky the FETs aren't damaged since the output is at the rail voltage. Resistors only fry when they can't dissipate the heat caused by the current flowing through them, so it would be wise to check any/all of the components in the circuitry associated with those resistors. If they were part of the input circuitry and there is an op amp in there you may want to check the op amp and any power supply circuitry for the op amp. |
|
|
| haakonkl |
I'm not very competent in electronics, but having just assembled my first ZAP 2.3SE amp, I have wondered about a question that is related to your problem.
The ZAP works by switching +/- voltage 400.000 times or so every second; the input signal controls the length of the +/- timeslices. When there is no input signal, and the output therefore should be 0, it is actually switching between + and - anyway, with equal timeslices.
What if the clock should stop? Hang up? Or, maybe be burned out? Would that draw a permanent voltage through one of the FET's?
Haakon. |
|
|
| john65b |
Hmm,
Or if only the +rail is active, - rail dead, and no input, would you see +rail voltage on the output? And vice versa?
I will check (hopefully sometime tonight if possible) what the exact voltage is at outputs are when +/- rail and input are connected.
It's been a good 6 - 8 months since I had my little accident. |
|
|
| zkaiser |
| I thought that we agreeded to post high resolution pic of top and bottom or else there is no way that I could finishup my sch extracting ?? |
|
|
| john65b |
| I am working on it...waiting for the good camera... |
|
|
| zkaiser |
| Ok, waiting until then... |
|
|
| john65b |
| Well these pics might be a bit better, but the good camera is not much better than the crappy one... |
|
|
| zkaiser |
| Sorry, still not good enough.... |
|
|
| BWRX |
If you're going to try and reverse engineer it why don't you at least buy a couple modules?
Are you doing this just for your own personal gain? |
|
|
| john65b |
| I'll sell you mine...cheap too. |
|
|
| zkaiser |
First of all, this whole thing is more or less for fun, second I don't need phisical object which is also epoxy sealed (if I have good pic there is no differnce to real one), and finally I don't have time currently to deal with these due to studies obligations....
but I wanted to help, if you just want to sell this as a cheap junk thats different thing ... |
|
|
| john65b |
| Hey, I just read of someone losing their feedback resistor on a chip amp and it put full DC on the outputs...is this possibly the issue with these modules? Where is the Feedback resistor anyway? |
|
|
| zkaiser |
Here they are R||R + C to gnd+ R -> - in of opamp.
Ps.
any chances of sharper image? |
|
|
| john65b |
SO I am reviving this thread - I am quite sure I mistakenly mixed up the Power Ground and Negative supply when I saw smoke.
As far as the MOSFETs, the following numbers are shown on my DMM in DIODE mode MOSFET #1 is to the right ao the V+ Side of the amp module)
MOSFET #1 GATE - SOURCE - 759
MOSFET #2 GATE - SOURCE - 731
MOSFET #3 GATE - SOURCE - OVERLOAD
Does this mean MOSFET #3 is likely blown, or maybe MOSFET #1 & 2 are blown? Doesn't seem to be a short anywhere on either MOSFET (none of the legs shorting to each other)
Also, These are IRF640N and supposed to be IRFB31N20, correct (ZAP V 2.1 Standard) |
|
|
| john65b |
upon putting +/-20VDC on the rails, I get the following:
MOSFET #1 Drain to Source -14V
Drain to Gate +14V
MOSFET #2 DRain to Source -24V
Drain to Gate +24V
MOSFET #3 Drain to Source 0V
Drain to Gate 0V
MOSFET #1 is on V+ side of module.
And I am getting a consistent 5V on the output.
Like I said, I am pretty sure I accidentally mixed up the -V Rail and Power Ground
Any clues? Anything look wrong? I just bought six IRF640N to replace some/all of the MOSFETS and should be here in a few days... |
|
|
| john65b |
OK, so I noticed a fried 1K resistor in front of the -V side cap, replaced it and now have the following:
upon putting +/-20VDC on the rails, I get the following:
MOSFET #1 Drain to Source -14V
Drain to Gate +14V
MOSFET #2 Drain to Source -24V
Drain to Gate +24V
MOSFET #3 Drain to Source -20V
Drain to Gate +20V
DMM in DIODE mode MOSFET #1 is to the right and the V+ Side of the amp module)
MOSFET #1 GATE - SOURCE - 614
MOSFET #2 GATE - SOURCE - 613
MOSFET #3 GATE - SOURCE - 615
But I still have +5VDC on the Outputs... |
|
|
| john65b |
Zkaiser -
In Post #19 I showed the SMD resistor that was fried - I had some one say 20k and someone else say 1K...
Do you know what it is?
Appreciate the help...never was able to get that hi res digital pic for you... |
|
|
| luka |
Hi
Send them too me, I will post them + I like those kind of pictures :D. My email is down here...
|
|
V |
|
|
| zkaiser |
I have this picture which leeds me to think that this resistor value is, let's say about 1.6ohm with 1% tolerance (black blue gold brown) which is very reasonable value to burn at this power rating (1/4W)...
PS
If anytime you find some way to shoot hi res pictures I will be glad to see them. |
|
|
| john65b |
This is the resistor I am referring to in red...
Looks like Black/Blue/Gold/Brown to me too...
1 Ohm? 1.6 Ohm?
Hmmm, I will try 1.6 Ohm and see if I get the offset still.
I have not forgotten about your pics... |
|
|
| zkaiser |
| this black ring was already there I didn't draw it, after all it is transistor |
|
|
| john65b |
Yes, a transistor - I didn't look closely at what was in the ring - just saw the ring in the wrong spot...
I see another resistor (R14 and R16) by the MOSFETs with color of Grey/Black/Blue/Gold/Brown.
It measures 82ohms on the board...I am really confused now...
Either way, a 1K Ohm or a 1.6 Ohm did not stop the + 5V Offset - it shows the exact same resistance.
As I said, I have +/- 20VDC on rails....which is lower than what they advise (+/- 32VDC is min according to data sheet)...could the lower voltage rails be a reason there is an offset?
FETS are not shorted, all major caps and all large resistors on the perimeter all seem to be OK too... |
|
|
| john65b |
Well I just put 35VDC on the rails and got -35VDC on the outputs. Blue LED came on.
Back to square one. |
|
|
| Lars Clausen |
John65b, i will try to help you if i can.
The resistors in question are 80.6 Ohms. So 82 Ohms seems to be fine.
Did you have a load on the output at startup?
Did you measure the voltage on the input op-amp? Pin 4 should be -4.5V and pin 8 should be +4.5 referenced to GND.
Also check the 2.2 Ohms power resistor at the input side for open circuit. It's a GND connecting resistor, so you can just use you ohm-meter from input GND to output GND, and check for connectivity ~ 2 Ohms.
I hope it helps. |
|
|
| john65b |
| quote: | | Also check the 2.2 Ohms power resistor at the input side for open circuit. It's a GND connecting resistor, so you can just use you ohm-meter from input GND to output GND, and check for connectivity ~ 2 Ohms. |
Input GND to output GND is 2.7 ohms, so OK here. |
|
|
| john65b |
| quote: | | Did you have a load on the output at startup? |
I initially had no load across the output and no load on input and powered up...full -35VDC on output
I put a 8 ohm load resistor across the output and still had around -20VDC at output. |
|
|
| john65b |
| quote: | | The resistors in question are 80.6 Ohms. So 82 Ohms seems to be fine. |
I currently have one module with a 1/4 watt 1K ohm resistor, and the other with a 1/4 watt 1 ohm resistor.
I can put the correct 80 ohm resistor in place, but wonder if the issue is somewhere else, since both modules are having DC on rails still with resistor higher and lower than the required 80 ohm....I will put it in anyway and re check the outputs.
Seems with this resistor out, the MOSFET is wide open |
|
|
| john65b |
| quote: | | Did you measure the voltage on the input op-amp? Pin 4 should be -4.5V and pin 8 should be +4.5 referenced to GND. |
Measured -4.8V on pin 4 and +4.85 on pin 8 referenced to input GND, so OK here too... |
|
|
| Lars Clausen |
And both modules are doing the same?
Both -35V out? |
|
|
| john65b |
Now that I have changed the burnt resistor on the one module from 1K ohm to 68 ohm (I did not have an 82 ohm available) the DC at output is more like positive 10V.
I can get an 82 ohm resistor and try. But still have no sound at the output with the 68ohm.
The module with 1 ohm is still at -35V on output.
|
|
|
| Lars Clausen |
what have you tied the gate drive supply to?
Have the modules been working in your setup at one time? |
|
|
| john65b |
The modules worked at one time.
I beleive I had misconnected the power GND and -V rail a while back. Saw some smoke, so I disconnected the module. The only damage I saw was the 80 ohm resistor in front of the caps. They were charred, so I removed them. I was unable to dterermine their value. I sent an emauil to LCAudio, and afetr a few emails they said it was 1K, but I knew they were thinking of another resistor.
None of the MOSFETS are not shorted to each other. Both square caps seem to be fine (show 22uf on my LC meter). the Rifa caps below also seem fine (caps not open, or shorted). On the Rifa caps below, the two closest to power/output connections shoe .22uF in circuit and teh back ones show 2.2uF in circuit)
What have I connected to the Gate Drive Supply? Sorry, I don't understand. All connections to the modules are +V, -V, V GND, Sig in, Sig GND, Out, Out GND.
I have no other connections on the modules. |
|
|
| Lars Clausen |
OK, it's likely the modules are damaged from the supply break.
You should measure the Vgate on the IRF640N mosfet. It should be around 12V. (On the Source). Be careful not to short anything with your meter. |
|
|
| john65b |
OK, measure the three MOSFETs. A MOSFET that is not shorted can still be bad. This is what I was hoping.
MOSFET Tab (or drain) to Source Pin should be 12VDC with no connection on input or output...Got it...will try tonight.
I just ordered six IRF640 MOSFETS the other day thinking this is what needed to be replaced. Hopefully this will be the issue.
Mr Clausen, I really appreciate your assistance.
I have a Tripath TA3020 Amp, a UCD400 amp and was hoping to get the ZAP working so I can do my own personal review.
Thanks again. |
|
|
| john65b |
Here are the voltages I get across the MOSFETS - all voltages are grounded to tab/drain
Module #1
MOSFET #1
Source -23.5V (+V Rail Side)
Gate -23.5V
MOSFET #2
Source -43V
Gate -43V
MOSFET #3 (-V Rail Side)
Source -22V
Gate -18.5V
Module #2
MOSFET #1(+V Rail Side)
Source -70V
Gate -70V
MOSFET #2
Source 0V
Gate +12V
MOSFET #3 (-V Rail Side)
Source -23V
Gate -19.4V
All six appear bad? Maybe MOSFET #2 on Module 2 is OK? |
|
|
| john65b |
After thinking about it further, I do not believe I miss connected the GND and -V rail, but either the GND or -V rail may have not been securely soldered onto the connection. Would this cause the problem losing the GND or a rail?
Anyway, the Voltages should read:
Module #1
MOSFET #1 (+V Rail Side)
Source -23.5V
Gate -23.5V
MOSFET #2
Source -43V
Gate -43V
MOSFET #3 (-V Rail Side)
Source -22V
Gate -18.5V
Module #2
MOSFET #1(+V Rail Side)
Source -70V
Gate -70V
MOSFET #2
Source 0V
Gate +12V
MOSFET #3 (-V Rail Side)
Source -23V
Gate -19.4V |
|
|
| Lars Clausen |
It's a bit hard to troubleshoot from these measurements, especially since they are taken from source to drain etc. and in uneven order.
Could you repost the measurejments, this time with GND as reference, and starting from the top V+ mosfet?
That would make it somewhat clearer i hope.
Thanks |
|
|
| john65b |
Lars,
Yes, I will retake those measurements from GND.
Sorry, and thanks again for the help. |
|
|
| john65b |
OK, once again Voltages referenced from Power Ground:
Module 1
MOSFET #1 (V+ Side)
Vs = +10V
Vg = +10V
MOSFET #2
Vs = -36.3V
Vg = -36.3V
MOSFET #3
Vs = -24V
Vg = -20.2V
Module #2
MOSFET #1 (V+ Side)
Vs = -36V
Vg = -36V
MOSFET #2
Vs = -36.5V
Vg = -24.3V
MOSFET #3
Vs = -24.2V
Vg = -20.4V
Hope this helps. |
|
|
| ozangazi |
| i think you must clear fet pins..Soldier relic is much unfavorable for low signal and high frequency applications.. |
|
|
| john65b |
| quote: | | i think you must clear fet pins..Soldier relic is much unfavorable for low signal and high frequency applications.. |
Clear FET pins? I do not understand.
I have received six new IRF640 MOSFETs. I can replace them if need be.
I also took these voltages with no connections to input and output. Was I supposed to have a dummy load across the output for these Gate/Source/Drain voltages to be real/accurate?
Sorry for all the lame questions.
I am beginning to think these modules are KAPUT... thinking of getting another UCD400 or finish my AMP2. But then again I need another amp ilke I need a hole in my head. |
|
|
| ozangazi |
i mean,clean solder paste remains after soldering ...
Sorry for poorly english :( |
|
|
| john65b |
Currently, there is no solder across pins - the voltages I have are from a stock, unmodified module.
I may replace the MOSFETs and will remember to keep them clean.
Thanks |
|
|
|
