Tubes in broken spotwelder & other questions

Round. Shiny. Get hot. They hold the magic smoke so well, it usually spurts out of some other part of the circuit when they break. What more do you need to know?

Probably 200uF, but I won't weep if I'm found to be wrong.

Charging the caps through the voltage regulators would be dumb, and may even overheat the tubes, so I'm not sure I believe it. Much more likely is that they're being used to generate a regulated 450VDC supply someplace (the OA2 is a nominal 150V regulator). I'd claim that they're there to protect the caps, but they have a strike voltage of 185V, three times which is greater than the voltage rating on the caps -- and smart engineers never run their 'lytics at full voltage rating.

Did you find this by measurement, or what? Industrial circuits do thingsn that I find weird so I could be wrong, but I would expect the chain of regulator tubes to end at ground.

Gas regulator tubes like that are basically neon lights, only they're not for making light and the gas mix may not have neon. The gas mix and the electrode spacing is selected for the desired voltage drop. IIRC, those will light up blue or purple if they're working -- but it's been a while. They'll light up with some visible color, although the room may need to be dim to see them. So dim the lights and look.

I would expect that if the circuit has thyratrons in it at all, that they are there for closing the primary power circuit, not for closing the relay. I could be wrong (and those are dinky little tubes, which makes me think I might indeed be wrong), but I'd only feel sure of that if I knew that the thyratron anodes were connected to the relay coil, rather than the secondary.

Have you typed the model information into Google and seen if any schematics pop up? This whole job is really best done if you know the schematic. I think that if I was as far along as you were in this I'd be spending some quality time with the welder, an ohmmeter, a pencil and paper, and tracing the schematic out by hand -- assuming I couldn't find it on the web.

You talk about hearing the relay click -- does that mean that you've had the thing plugged in? If you've had it plugged in and smoke didn't pour out, then it's passed the first test and you can probably safely play with it (well, as safely as you can do any probing around in a circuit with at least 450V on it).

Yes, you can test the caps in the way you suggest, at least to some extent. The one that rattles is probably dead. Charge each one up and note the current it draws when charged up (Google "capacitor reforming" if you want to do the whole dance). Then, if you have a storage oscilloscope, discharge each one through a known resistance and look at how quickly the voltage drops (or discharge one through a big enough resistance that you can watch on a voltmeter). Assuming that they'll hold charge for a minute or so by themselves, when you put a resistor on the amount of time that it takes them to get from full voltage to 38% of full is the RC time constant of the circuit -- if you know the resistance and the time, you can solve t = RC for the capacitance.

If you've had the thing running, and you know how to be safe around voltages like that then I suggest you trace out the schematic, or try to obtain one, then test things in operation (Google for procedures -- 450V can leave you dead before you realize it, and I don't want to be reading your obituary).

Do you know anythng about electronics at all ? Maybe the solid state electronics.

The capacitors marked 200 450 mean they are 200 mfd and rated to run at up to 450 volts on them. They don't have to have that much, just the maximum.

The 0A2 tubes are like big neon bulbs. The show glow a blue color when working. You can look up the voltage as I don't want to. Somewhere around 100 to 150 volts will be required to make them glow. When glowing they regulate the voltage to that value. Only 2 pins are needed for them.

You can get an idea of the capacitors if you have an old analog ohm meter or multimeter. Go to a high ohms scale and see how if the ohms show very low to start with and after several seconds the ohms should go to almost an open circuit indication. This will probably have to be done with atleast one side out of the circuit.

The 5U4 should have a filiment you can see glowing. There are 2 pins connected to the 2 plates. You should get pulsating DC from each plate to the ground or common connection point probably where the negative lead of the capacitors are. If the capacitors are good or sort of working the DC will be steady.

Oops -- I pulled up the 2D21 data sheet but only noticed the following when I went to close that browser tab: " ... designed for use in relay applications ... ". So it could just be there to make the relay work.

I'm a bit puzzled, because unless that's one honkin' big relay, it seems like the capacitor bank is too large to be consistent with just turning it on, but the power transformer is too damned small to weld with. I'm leaving all my ramblings above intact, but check if the welding transformer doesn't just go straight to the input power -- I'm wondering if all the fancy electronics aren't just there to set the timing on the pulse, and the actual welding power is AC (why not?) and steered by the relay.

Bwahahahahaha! Good one! ...Jim Thompson

-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | STV, Queen Creek, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

I'm looking for work... see my website.

The relay works-I saw it working. And it does indeed connect the cap bank to the output xmfr. The connections are obvious. And the relay contacts are big, like a contactor.

Thanks for the reply Ralph. Especially the info about the large caps. I do OK with solid state stuff but tubes are a mystery to me because I have never really needed to work with them or test them. My son has googled for a schematic but so far no luck. It may be that he will need to trace it out himself. Eric

The OA2s are like zener diodes and they are 150V. I would say they have 3 in series to get a regulated 450V to charge the caps, BUT then why are the caps only rated at 450V? I've been working on a device that has an OA2 and an OB2 in series for a regulated 258Vdc. Must be the OB2 is a 108V regulator. Just my 2 cents to add to the confusion. Mikek

If it helps your thought processes... pentodes are similar in function to depletion-mode MOSFETs, triodes likewise, but with lousy LAMBDA ;-) ...Jim Thompson

A thing to be wary of... I got bit by it when I was still a teen... gas discharge regulator tubes may fail to fire when stored in the dark for a long time, resulting in too high a voltage applied to electrolytics with resultant loss of the magic smoke ;-0 ...Jim Thompson

You couldn't pay me enough to spot weld with that thing.

And thyristors are kinda like SCR's.

Eh? Thyristors/SCR's are "both" semiconductors. Maybe you meant to say thyratron ?? ...Jim Thompson

Thyratron, yes. Oops.

Why? It looks well built, if a bit old.

It's probably stepped down considerably from that.

The caps discharge into a large transformer. The spotwelding voltage must be very low, just a few volts. ALL of the spotwelders I have ever used were high current, low voltage. Usually around three volts. Eric

Yes, If the unit is working properly.

If that's your reasoning then don't use any spot welder.