Power electronics schematics - Cebora 816

[pointer2null]

Got the mig all set up and started having a go. Think I'll need a lot of practive

Anyhow, after about 5 minutes of burning holes in my test scrap etc a problem developed with the mig. It sounds like the system is under load. I've had a good check for bad connections and the like, but all seems fine. When the noise starts I've notice the gas flow rate meter dips as if the gas solonoid is being operated. To cut a long story short, i think I've got some form of short circuit somewhere which is activating the mig - not continuosly but "pulsing" so the noise is actually the solonoids buzzing on and off - the sort of thing you'd get if you put a dc relay on ac.

I've got a schematic of the welder but I'm more used to micro electronic schematics and I'm not sure of all the symbols on this one:

I've labeled all the bits I know and the two I'm not sure about: ??? 1 and what looks like a cut out switch. Any ideas?

These are the bits in real life:

As you can just about see the main solonoid had a manually operatable button in the middle, when the noise starts this vibrates like hell and slowly pulls in.

Also - anyone know how to read the rotary switch diagram?

 

[Jake SS]

Sounds like it might be a lack of voltage to close the contactor and gas valve?
They would probably be 24v supply to the coils of both.

The rotary switch has an "off" = "0" then 1-6, in position 1 = A-1 is closed (contacting), B-2 is closed, R-3 is closed, S-7 is closed.

In position 2 = A-1 is closed, B-2 is closed, R-3 is closed, S-6 is closed. And so on through to number 6 position.

???1 = Supply to light,fan and auxiliary transformer when A-1 and B-2 are closed (positions 1 through to 6)

Black dots mean that the contact numbers to the left are closed, top numbers are switch positions.

Hope that makes sense.

Think maybe the "cutout" is the trigger, looks like a push button symbol? Trace the wires.

 

[pointer2null]

I think I've got it all figured now. Have to remember this diagram is italian and from 1984 so some symbols have changed (I guess).

If I'm right on what is what then my problem looks to be on the PCB. Trying to get a schematic of that now.

 

[R Kraft]

You seem to be getting low voltage applied to the main relay.
You may have a dirty torch switch or if it has one, a problem with the small relay on the PCB.
If you have a meter you could check the voltage across the gas valve to see if it is getting the rated voltage.
Robert

 

[eddie49]

When the torch trigger is pressed it will signal the control pcb to start the wire feed motor and pull in the main CB and the gas valve. These latter two may be swiched through a power transistor on the pcb, or by means of a small relay. If the CB and the gas valve are chattering, it may be due to a faulty bridge rectifier on the pcb, which is passing half-wave rather than rectifying the 24 volt supply. That could cause the pcb to lose control, so the outputs are on even tho' you haven't pressed the torch trigger, and the half-wave power is pulsing the components. You could also try taking either the coil of the Circuit Breaker or the coil of the gas valve out of the circuit ( since they are in parallel ).

 

[shenion]

Could be rectifier or more likely a dead capacitor on the board.

 

[pointer2null]

Thanks guys.

I'll dig out the AVO later and have a nose around.

There is a relay on the PCB but I checked that and it wasn't operating when the fault occurred. There is a largish power transistor sat there looking guilty though, so I'll check that out as well.

May even have to dig out the old oscilloscope...

 

[pointer2null]

It's not the trigger switch - proved that by unplugging the torch and the problem persists.

Had a look over the PCB and it is very old technology. They've built the motor controller from descrete componates - op amps (4), comparitors (5) and silicon controlled rectifiers (2). You can get a chip for 50p now that does all that and more.

Anyhow, looks like the relays are controlled by a triac (3) and not a transistors. It's driven by a small opto isolator (1).

Without a schematic or just replacing componates, it's going to be difficult to repair.

Can you get generic boards for MIG's? All it really needs is the motor speed controller - I suppose as a last resport I could use a motor controller from Maplin... eek, never though the day would come when I'd say that.

 

[Rosepuds]

hi. This PCB is easy enough to repair. The triac and opto do energise the contactor and solenoid in the machine. These 2 components are the most likely to of failed but there is a possibility it could be another component that drives them or the snubber capacitor across it. I work for tecarc and repair PCB`s. PM me if you require further info.

 

[Jake SS]

Have you checked any voltages yet? Inputs / outputs etc.
The coils look as if they are AC supplied?
Check the thermal, bridge it then see what happens.
Go over PCB with Magnifier and check all joints and tracks.
You really have to know what voltages you are getting.

Cheers

 

[pointer2null]

I've tried shorting the input to the opto coupler when the fault occurs. This stops the problem immediately so that rules out the opto and trac.

I've hear back from Cebora. The machine is almost 30 years old, but they still has the PCB schematic

The solonoid and gas valve are AC (makes sense as you can't switch DC with a triac)

So at the moment I'm hoping it's either supply caps/regulator putting noise on Vcc or even simpler - conductive dust trapped under one on the IC's. I'll see if I can check the power supply and on board regulator - the caps could have degraded.

 

[Jake SS]

You are starting at the end and reversing !
Try starting at the front and work your way to the end.
You will find it much easier to fault find.

 

[pointer2null]

I'm trying too lol

At the moment I'm staring at the schematic trying to work out whats going on - e.g. where is exactly does the torch switch connect in. Which bit is the PWM motor control etc.

 

[pointer2null]

I've ruled out the triac and opto - shorting the opto inputs stops the problem.
I've also rules out the driver transistor - shorting it's base (A) to ground stops the problem.
Grounding the top of diode D also stop the problem.

I've also rules out the LM324 (1) and the 15V zener (Z) - I replaced them.

I'm now thinking about side stepping the problem and bypassing the suspected circuitry. I won't be using the stitch weld functions which seem to controlled and generated by the op amp in the bottom right of the schematic (1), so may just try and disconnect it from the circuit (lift the diode on it's output that feed the inverted input of IC2.

Anyone know if a) this will work or b) have any other ideas?

 

[Rosepuds]

check the power supply. make sure you have a smooth, stable 24v dc. check diodes for being short or leaky. Does the wire feed motor try to turn aswell or is it just the contactor and solenoid buzzing.

 

[pointer2null]

Motor des not run, just the contactor and solonid buzz.

I'll double check the on board power rail - dug the scope out now.

Also just got back from a mate who happened to have a spare opto isolator - will also try that just in case.

 

[pointer2null]

Check the 20V power rail - it's at 20V and stable - no sign of noise or AC on the line.

Swapped the opto (put a 3021 in) and the fault persists.

 

[Jake SS]

What is the voltage at the coils of the contactor and/or valve when triggered?

 

[pointer2null]

I think I've got it - replaced the diver transistor to the opto and it looks like it's ok now. I should have got it before, but there was no signal visable on the scope on the pins of the opto.

Thanks for everyones help and tips. Fingers crossed that transistor was the culprit.

 

[pointer2null]

Just had a little play and did my first real welds and the mig behaved itself perfectly. Sorted