Any circuit board gurus lurking, diesel welder wont come off high idle.

[chrismac]

On measurement 1 the value is constantly changing, I was going to state that when I typed that earlier.

The lamp certainly remains lit for the period I have left it set up for, which was certainly a few minutes,

In the welder with the solenoid isolated yes, the revs just drop off back to idle.

 

[chrismac]

+12V (Battery +) to B
Ground (Battery - ) to J
Lamp between B and C

Not sure what 12V supply you've got, but If you're thinking of running this of a big 12V lead acid battery it might be prudent to put a fuse inline with the +12v, just to save from any mishaps should you accidentally short anything out.

Once that is rigged, then you want to connect D to B this should simulate you striking an arc.
The lamp should light at this point because it's pretending to be the solenoid.
If the board is working then we wouldn't expect the lamp to turn off until a few seconds after you remove the D to B wire.

However, based on what you've observed, we think the board will turn the lamp off after 5 seconds.

I am going to try this set up on the machine with the solenoid insulated and uninsulated form the body of the machine out of curiosity to see is the results change.

 

[rich_c]

What did you mean by "try this setup" with the solenoid, I'm not sure I fully understand? How much of this rig are you connecting to the welder.
Note of caution depending on how you're about to hook stuff together... be aware that when board is in the welder and fully wired Pin D is connected to the starter switch and starter solenoid. You don't want to be shorting B to D if you also hooked up the original D wire too.

I haven't had a good sit down with the circuit to think through all the implications of your measurements yet, but we need to get to the bottom of why you're seeing different results between bench and welder - I think at one stage you saw CB1 repeatedly tripping and resetting, but you said you'd fixed that issue right?

 

[chrismac]

What did you mean by "try this setup" with the solenoid, I'm not sure I fully understand? How much of this rig are you connecting to the welder.
Note of caution depending on how you're about to hook stuff together... be aware that when board is in the welder and fully wired Pin D is connected to the starter switch and starter solenoid. You don't want to be shorting B to D if you also hooked up the original D wire too.

I haven't had a good sit down with the circuit to think through all the implications of your measurements yet, but we need to get to the bottom of why you're seeing different results between bench and welder - I think at one stage you saw CB1 repeatedly tripping and resetting, but you said you'd fixed that issue right?

I'd be curious to know the no-weld, and weld voltages at the following points if you can manage it. (only need measurements in auto-idle mode.)

1. The other side of R56 (junction of R56, R57, C53 and Q51-A)
2. The gate voltage of Q51 (junction of D53, R63, and Q51-G)
3. The gate voltage of SCR51 (Junction if R64 and R60)

Those are my top 3 to try and learn something about what's happening, but anything you care to probe is probably interesting too.

If it was me I'd be tempted to pull the board out and test it on a bench so you don't have to keep engine running and can have nice sit down and a cuppa. You'd just need a 12V supply, and a load (a 12v lamp would do). You could then simulate welding current by touching spade D (or E) to 12V.

The tests above were carried out on the bench, I was going to try and get readings while the machine was actually in idle and welding mode (arc being made) with the solenoid insulated and bolted direct to see if there was a difference,

Yes the issue with CB1 being over loaded has been rectified, a sticky solenoid plunger, or very dry and when being pulled down at an angle was binding before full travel was being achieved, meaning 20amps pull was not dropping to less than 1amp.

I appreciate the time your taking to go through all these test results.

 

[rich_c]

If you putting it back on the welder to take measurements, one that you could add would be the Anode of SCR51 which is available on spade terminal A too.
Because we're into time based measurements, the before welding measurement, during welding and then the moment it drops out of welding. My prediction is that before welding it will be near 0V, during it will be near 12V, and when it drops out of welding back to 0V.

If you want to do a test similar to the shorting B to D test while on the welder. Then short B to E. That would give us insight into whether there is something weird going on with the current transformer input to the board.

 

[chrismac]

If you putting it back on the welder to take measurements, one that you could add would be the Anode of SCR51 which is available on spade terminal A too.
Because we're into time based measurements, the before welding measurement, during welding and then the moment it drops out of welding. My prediction is that before welding it will be near 0V, during it will be near 12V, and when it drops out of welding back to 0V.

If you want to do a test similar to the shorting B to D test while on the welder. Then short B to E. That would give us insight into whether there is something weird going on with the current transformer input to the board.

Spade terminals A and E are currently unused, to confirm the test on B to E is giving input voltage so the solenoid should become energised and pull the revs up, if working good it should then hold for 15seconds as per manual and drop the revs back to idle.

To help me understand your prediction, machine all set up on weld mode, auto idle on (revs high when solenoid is hard fixed) in this set up it could be 12v continuously due to the fault we are seeing,
and then weld mode, auto idle on (revs low when solenoid is insulated from body) in this set up it could be 0v pre weld, 12v when arc struck then dropping to 0v when the revs drops on its own while welding.

 

[rich_c]

Yeah, A, E are not wired into welder. But according to the schematic should be connected to SCR51-A and RESET on the PCB respectively.

B to E is essentially the same as B to D but without the side effect of energising the starter motor solenoid!

I feel I should read the manual in detail, my assumption from what you and/or eddie had said from the manual was that revs would hold for the duration of B to E contact (or duration of welding), then after removing short (or stopping welding) it would take 15 sec delay before dropping back to idle.

 

[chrismac]

I was going to leave it a day or so but as the board was on the bench I thought I could get some reading from it with your last request, so the Anode (centre pin) of SCR 51 to pin J, rigged up with the battery and lamp as before,

when replicating not welding voltage up and down not sitting steady or at zero, see video.

please let me know if you can view the video..

https://flic.kr/p/2p6JyQi

then when replicating welding pins (D & B linked) steady battery voltage approx 12.3V,

 

[rich_c]

Yeah, I can see the video. Hard to tell as the multimeter can't keep up, but it looks like it ramps up from near 0V to near 12V and then instantly back to zero. Repeating every half second or so?

On the bright side it is a constant 12V when welding, which is expected

Next thoughts:
1. Desolder one leg of R57 and check that it's resistance is roughly 620k
2. If you can measure capacitance check C53 is 22uF (desolder 1 leg). Or replace with known working 22uF capacitor.

 

[chrismac]

1. I'll check the resistance of R57, its out on its own so hopefully will come out without too much drama, the board has been coated in a protective lacquer, its a bit of a fight sometimes to get the components loose with out damage,
2. C53 was changed in August (that was a post I found on another forum as a possible fault ) the old one did not show continuity when removed which I hoped would mean it was functioning, I dont think I can check capacitance.

I stand corrected my meter can test capacitance. I'll give that a go and see how it preforms.

 

[chrismac]

Yeah, I can see the video. Hard to tell as the multimeter can't keep up, but it looks like it ramps up from near 0V to near 12V and then instantly back to zero. Repeating every half second or so?

On the bright side it is a constant 12V when welding, which is expected

Next thoughts:
1. Desolder one leg of R57 and check that it's resistance is roughly 620k
2. If you can measure capacitance check C53 is 22uF (desolder 1 leg). Or replace with known working 22uF capacitor.

Managed to get one leg of the resistor R57 in the centre of the soldering hole as I could not get it removed due to the lacquer, got 0.625m ohms = 625k,

Then the find of the night my ability to test capacitors, C53 had been replaced by me, but it had failed, I put the old one back in as it was testing good, left it until tonight when I thought I should have carried out the lamp test, home after work and set up the lamp like before, low and behold, it does what its supposed to, on and off after approx 10-15 seconds after removing the wire link between B and D

I fitted it on the machine with the solenoid fixed direct as it should be, same as before stuck on high idle revs. insulated solenoid and it works perfect, it must be the solenoid which has failed internally some how grounding to the body.

Thankyou to everyone who had input, @eddie49 in the early days and @rich_c the time you spent looking over the circuit diagram sussing out what does what was invaluable in guiding me to do what obviously to you it bread and butter, the bench test was perfect to confirm the board was working, I can only thankyou.

 

[rich_c]

Phew, got there in the end, excellent news

Did you spot that C53 is polarised, so needs to go in the right way round, otherwise it will work for a while then fail some time later.

 

[chrismac]

Yes spotted that thanks, not sure what I'll do with all my free time now, probably do what I should have been doing the last few weeks instead of working on this