Maker Pro
Maker Pro

PCB not functioning

QuantumCheese

Apr 27, 2012
74
Joined
Apr 27, 2012
Messages
74
partyanimallighting; said:
I basically treat it like an ex-wife. .
LOL

Hi again, good to hear progress is being made.

on to the board (#1),
if VR2 was allowing a -ve voltage to get to that component it's probably already too late. and worse than that, yes you could buy another one, but i bet you won't be able to extract the code from the working one to program up a replacement. I'm more familiar with PIC's but i'll bet that family of uProcs has similar code protection.

If your absolutely sure the power supply (5V) is now correct you could try swapping U3 over from the other board, but bare in mind if it's not fixed it could blow the known-good chip!
 

partyanimallighting

Oct 22, 2012
330
Joined
Oct 22, 2012
Messages
330
Here's the plan and let me know what you think. It's possible that, if I swap out U3 from the working unit, I could blow the good chip from Board # 2 in Board # 1 that has the problem. But.......and tell me if I'm wrong here. If I take out the possible bad chip from Board # 1 and put in Board # 2 and it doesn't work, it's obviously blown. But if it works???? That would tell me there's another problem on Board # 1. I'm getting the lamp to flash when I short SCR-22 and MOC8104 so I was just wondering...... What do you think?
 

QuantumCheese

Apr 27, 2012
74
Joined
Apr 27, 2012
Messages
74
Brilliant idea batman! why didn't i think of that??

Shorting the pins on the Opto-isolator (MOC8104) is effectively sending the same signal to the HT side of the board as the uProc does. so by doing that you have proved the HT side works fine.
I'll hang fire on speculating what to do next until you've tried the suspected duff U3 in the working board.
 

partyanimallighting

Oct 22, 2012
330
Joined
Oct 22, 2012
Messages
330
Hi QC, I pulled the possibly defective U3 from the defective pcb (board # 1) and installed it in the working pcb and.......the lamp flashed!!! But!!!!.......it flashed really, really dim. I changed out the lamp with a new lamp and the result was the same, dim flashing. I then changed back the U3's, re-tested the working unit and it's fine. My question is, obviously the U3 is faulty, but I still have to get the defective pcb to actually flash the lamp so there's another issue elsewhere on pcb # 1. Feedback, anyone???
 

QuantumCheese

Apr 27, 2012
74
Joined
Apr 27, 2012
Messages
74
A 'dim flash'? that is curious.

If you remove U3 (from either board) then carefully short socket pin 12 to gnd (tab of vr1 or2), that should trigger it to flash on either board and at the same intensity.

Anyone else wants to jump in here, please do as i'm having a little difficulty working out how a 'dim flash' is possible

When it flashes 'dim' is it at the same frequency as a 'normal' flash?
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
I'm having a hard time getting to that 4shared folder with the schematics.

Is this a xenon flash?

If so, check the capacitor. If the capacitance is significantly reduced, that could be the problem (or if it is charged to too low a voltage).

OK, finally got it.

I would monitor the voltage across the capacitor, then see how fast it drops after the power is removed. Compare the faulty device with the good one.

If the voltage falls MUCH faster on one than the other, this indicates a problem.

I'm not sure how likely this is, but it's worth a shot. Remember to connect the multimeter before power is applied, and don't touch it until everything is safe, disconnected and discharged again.
 

QuantumCheese

Apr 27, 2012
74
Joined
Apr 27, 2012
Messages
74
the things is, i would be thinking along those lines too it's just that it apparently the flashes normal intensity with one uProc and dim with a suspect one.
I think it might be somewhere in the feedback i'm just not sure where!
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Yeah, I read it as U3 was faulty, but even with a good U3 he still got a dim flash (indicating another fault on the board).

But on re-reading it, the dim flash moves with the faulty U3.

Could it be that it triggers the flash before the capacitor is charged fully? (seems unlikely). As I see it, any trigger pulse strong enough to trigger the flash tube is going to result in a full flash. (But maybe I'm wrong)

But clearly something is going on...
 

QuantumCheese

Apr 27, 2012
74
Joined
Apr 27, 2012
Messages
74
Yea i'm not sure either! I'm wondering if U3 is totally shot and its i/o is in a cmos-dead-unknown-state which is just low enough to trigger the opto into firing it's output but not hard-on leading to q1 (the SCR) not firing full on either.

I'm hoping then when partyanimallighting manually triggers it on the opto all will become clear....
 

partyanimallighting

Oct 22, 2012
330
Joined
Oct 22, 2012
Messages
330
OK. So far, test to see if the flash rate varies with the different U3's in the working unit. Then short pin 12 on U3 socket on both pcb's to ground and check intensity. It should be the same intensity, right?
On the issue of testing the capacitor, I can't check the drop on the faulty pcb because the lamp does not flash on this unit so I'll need some advice here. Do you want me to check the voltage drop regardless of whether it works or not? And you're referring to the larger cap? There's one by each reg.
My "novice" view is that, if U3 from the defective pcb was placed in the good pcb and the flash was dim, obviously this U3 is faulty. But that still does not solve the issue with the defective pcb not flashing at all. As soon as I run these tests I'll post. The weekend is upon me and I'm up to my wazoo in jobs so I won't be able to test until probably Monday. Keep your fingers crossed for me. I can get replacement parts from the company so that's a positive to all this. Thanks for all the info and assistance.
 

partyanimallighting

Oct 22, 2012
330
Joined
Oct 22, 2012
Messages
330
Hi guys, sorry for the delay but I was in the island of Tobago for one week then, when I got back, got the flu. Here are the results as requested. I replaced LP2954 with LM7805 as this was defective. I hope this would not affect the test results. I re-tested both U3's again in the working unit and the flash rate on the U3 from the defective unit has the same problem, it flashes very dim. The strobe units themselves work on two channels, Channel 1 is Flash Intensity and Channel 2 is Flash Rate. I'm getting full control, intensity is adjustable, flash rate is fine. It's just that it flashes very dim. I then removed U3 from both PCB's, shorted pin 12 on U3 to ground and got the same result on both units, a full bright flash but when I short pin 12 on the defective PCB continuously, the flash dims considerably after a couple repeated shorts. Is this a symptom of another problem?
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Hi partyanimallighting,

OK, I've looked at the schematics and the design is pretty straightforward (although the schematics are not very clearly drawn). I will describe the general design, starting with the universal power supply schematic.

AC mains voltage enters at the top left, on "HOT" and "NEU" and is filtered by T1 and the capacitors, and appears across CC1.

The voltage feeds down to T3, which provides two low-voltage rails for the digital board. Pins S5 and S6 feed a bridge rectifier that feeds VR2 which regulates the +5V rail. There's also a feed from C11 called VOLT S01 which is used by the MCU in a clever way (I think) to detect the mains voltage.

T3 pins S7 and S8 feed half-wave rectifier D10, which supplies VR1 which generates the +12V rail, and also D9, which provides a mains-frequency signal into U4 on the logic board. (This voltage is generated across D8.) I suspect the MCU uses this for mains frequency detection, although I can't see how this knowledge could be used by the MCU.

These supply rails are all referenced to "GND" (indicated with the standard earth symbol near C13). This may or may not be connected to mains earth, which is shown with a different symbol, which I believe indicates a connection to the metal chassis, if there is one...

Going back to the direct AC mains voltage across CC1, this voltage is rectified by D4 (the large stud diode) and smoothed by C2 (with bleeder resistor R8) to provide the main supply rail for the strobe light. It also feeds a voltage doubler comprised of C1, D2 and D3, which supplies the firing circuit which is based around Q1.

Current via D1 and R3 and R2 charges C3 up to about +120V (limited by D5). When U1 receives an activation signal (current pulse) from the logic board, it feeds the voltage developed across R5 into the gate of Q1, a small SCR, making Q1 fire and conduct. This causes C3 to discharge rapidly into pulse transformer T2, whose secondary generates a very steep pulse of over 1kV peak which is connected to the trigger electrode on the Xenon flashtube. The pulse triggers conduction of the Xenon in the flashtube, which then converts the energy that is stored in C3 into a burst of bright light, discharging C3, which then recharges through R3 and R2.

The amount of power dissipated by these resistors depends on the voltage across them, which I believe will be significantly affected by the mains voltage, and their actual resistance (15K+15K or 22K+24K).

The digital board runs from the 5V and 12V rails generated on the power board. These come from T3 and (according to the schematic) are isolated from the mains. The 0V rail MAY be connected to mains earth; I'd like to know if it is - can you do a continuity check?

In this post I'm going to ask you to do a few things and answer a few questions; you may want to start a list :)

Hi Guys, if anybody's out there, I got unit # 2 repaired. Resistors R2 and R3 (22K/2W and24K/2W) were open and the pcb was a little scorched also. I replaced them, tested and the strobe works fine.

Me again. The LED was blown. Replaced it and the unit's fine. Someone please advise me as to the scorched parts on the pcb. I re-soldered everything back pretty good but please let me know if there's any cause for concern here.

I can't see a clear scenario where the resistors might overheat if nothing else was wrong. Has any component got bright solder on it, indicating that it was replaced, or at least resoldered, recently? I'm especially thinking of Q1, because if it failed, R2 and R3 could burn up.

It's interesting that this power supply claims to be "universal", but the voltage applied across the Xenon flash tube differs at different mains voltages.

I also noticed that R2 and R3 are shown on the schematic as both 15K, not 22K and 24K. I think there could be other variations from that schematic in the component values and types present in the unit(s) you have there. Can you find all the components in the high-voltage direct-mains-connected section and check that resistors and capacitors, especially R1, C1 and C3, are marked the same as on the schematic? And can you check that D1~4 are connected as shown in the schematic? That could help me be more confident that the schematic matches what you're working with.

I'm also interested in SW1 which is marked as a "THERMSW" (thermal switch I suppose). What can you tell me about it? Can you find it and take a photo of it? Also some photos of both sides with higher resolution would help, and you might want temporarily remove the flash tube socket for an "above" photo. You can email them to me if they won't fit here.

So your mains supply is 110VAC 60 Hz.

What voltages and other information are marked on the nameplate on the product you are using?

Do you know whether there are two types of the xenon bulbs, that are designed to work at different voltages? From my understanding of the circuit, the volage that this power supply delivers to the strobe lamp should be twice as high at 230VAC compared to 110VAC; in other words, it's not universal.

This might explain the problem with the unit that's flashing dimly. It may be designed to run on 230V not 110V. This is just a possibility that you can investigate.

Is it possible that the unit that had R2 and R3 burnt up had been used at a very high flash rate for a long time?

To all concerned, I am very very careful when I'm testing this pcb. QC made the dangers of working on this type of pcb very clear to me and I am very careful, switching off and unplugging before I go anywhere near it. I basically treat it like an ex-wife. You don't really want to have anything to do with it, but you have to!! I'll post some pix next.
Yes, many traces on the power supply board carry high voltages and are directly connected to the mains, so they can kill if you make any kind of contact with them.

As well as your good idea of securing the board firmly in an insulated jig, and working with one hand in your back pocket, you can improve your chances of not being electrocuted while measuring a live board (which is potentially very dangerous and you must only do it if you wish to accept that risk) is to solder short insulated wires with loops in the ends, onto the critical tracks (nodes) in the design. Never push a multimeter probe onto, or towards, the board, or anything live. If it slips, your hand can easily touch something live, and that can kill you.

At the moment I'm not asking you to make any measurements.

Also I'm confused by which unit we're talking about. Can you summarise each one in terms of what was wrong with it, what you're replaced and why, and what its current status is.
 
Last edited:

partyanimallighting

Oct 22, 2012
330
Joined
Oct 22, 2012
Messages
330
Hi Kris, thanks for the detailed reply. As it stands right now, I'm down to only one more defective board. I repaired the second problematic unit but have not re-assembled it as yet so I have this working board available for comparison. The repaired board was powering up with the Green LED lighting, but was not flashing (the red LED flashes in sync when the lamp flashes) and I found two open resistors (R2 22K 2W) and (R3 24k 2W), replaced them, and this unit worked fine.
On the problematic board, this was not powering up at all, no green LED, no function. I found LP2954 blown, replaced it temporarily with LM7805, and the unit powered up, green LED lit, but the lamp does not flash. As I mentioned above, I also swapped out U3 into the working board and the lamp flashed dimly.
I'll read your post thoroughly and get back to you with all requested information and photos.
 

partyanimallighting

Oct 22, 2012
330
Joined
Oct 22, 2012
Messages
330
By the way, is this a decent isolation transformer to buy to test these types of pcb's?
 

Attachments

  • 647_5_.jpg
    647_5_.jpg
    10.8 KB · Views: 104

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Presuming it is what it says on the box, that's very definitely an isolation transformer. Probably better and more expensive than you'd need even!
 
Top