Hunting HiVolt Power supply - No output

[Sr28]

I recently picked up a Hunting HiVolt Series 1000 power supply at a closing down auction rated at 0-500V, 0-1.5A.

After lugging it home, putting a fuse in and turning it on it made a very satisfying thunk and then sat and hummed. Unfortunately this was as far as it went as no matter what I have tried I cannot get it to produce an output.

From testing the transformers all appear to be outputting reasonable values and the main caps seem to hold a charge. The unit seems to be suck in current limiting mode but no current is ever displayed on the output. I probed the thyristor bridge rectifier and it is only producing about 10V across the output but after manually firing the thyristors it does seem to be generating a DC voltage and the caps charge up as normal. There does seem to be a voltage on the output terminals at this point but it is completely unregulated and does not display on the front panel either.

This leads me to believe there is some error in the sensing circuitry but I have no idea where to start! Non of the components look like they have let out the magic smoke and there are no burn marks. I haven't got any manuals and I cant find anything online either, any help is much appreciated!

 - Sam

[jeroen79]

Without a schematic I would start with the current sense resistor and the current limit potentiometer.
What voltages do you measure here?
Is current flowing or not?
Can you trace them to an opamp that compares the set limit to the actual current?
What do you measure on it's inputs and output?

[Andy Watson]

What is wired (or not wired) on the back panel ? The light (switch ?) on the front panel suggest there is an external sense circuit - has this been wired? 500V at 1.5A is lethal (very!) - I wouldn't be surprised if there is an interlock loop that has to be connected to enable the output.

[Sr28]

I haven't been able to work out what the switches on the front do or in which position they are active as the lights on them go out when they are pressed in. I have included a picture of the connections on the back plate. Pin 3 is connected to the positive output and Pin 12 the negative. Pin 11 is earthed to the chassis and Pin 10 is unconnected. All the other pins match up as the front panel suggests. I have tried feeding various voltages into the connector with the switches in all positions and this doesn't seem to have any effect. There are 2 other wires on pins 6 and 9 which I can't identify but they are Orange/Red and Red/Green. One appears to also go to the external sense switch and the other goes directly to the main board.

I tried connecting a high resistance load to it and again forced the thyristors to conduct and the reading on the current meter jumped up but I think this is just because it's wired in series with the output.

I managed to measure the + and - 12 rails for the Op-Amps and they are fine and the potentiometer for voltage measures 2.5v when set to the middle value. This might be a misunderstanding on my part but the current potentiometer always seems to measure 11.2v no matter where it is set.

 - Sam

[Sr28]

Continuing with the idea that there might be some sort of interlock or error in the sensing of the output I thought I might try connecting up the sense wires to the output. When turning it on again the output immediately started displaying the exactly 100V. Unfortunately this lasted about 3 seconds before the fuse blew. To make further testing easier I have bypassed the fuse and installed an inline MCB rated for 16 amps and inductive loads.

When turning on again it no longer displays a voltage and within seconds trips the MCB again. I've traced this back to the 2 silicon rectifying diodes in the bridge rectifier as they are both displaying a short circuit. I am going to pick up some more tomorrow and carry on testing.

Did these diodes die because they are from '87 and this is the first time in a long time this has actually been turned on or is it suggestive of a bigger problem like blown output transistors drawing too much current?

 - Sam

[Andy Watson]

I think you need to bite the bullet and trace out the circuit. Randomly replacing components and hoping for the best could be expensive.

The thyristors are obviously controlling the AC power through main transformer. I would guess that they are on the primary side and that they are controlled by the smaller of the two PCBs (the one with the two pulse transformers). Any imbalance in their operation will effectively lead to a large DC component and hence saturation of the main transformer core - and hence huge currents and many blown fuses.

Edit: That wiring loom is gorgeous! Not seen anything like that for years. The minimalism of the back panel makes me think that this unit was a special or adaptation of a standard module for a custom job, i.e. no chance of finding a service manual.

[Cerebus]

That wiring loom is gorgeous! Not seen anything like that for years. The minimalism of the back panel makes me think that this unit was a special or adaptation of a standard module for a custom job, i.e. no chance of finding a service manual.

Pretty common for quality gear of that era. A spool of lacing cord was a toolbox essential. The only problem is that kind of looming can play merry hell with you sometimes thanks to all the capacitive coupling it generates and the huge inductive loop areas it can include. Sometimes a messy looking layout can be electrically cleaner, but you try telling that to an assembly department supervisor that takes pride in doing a 'proper job'.

[Sr28]

I finally manged to get a chance to start tracing the circuit out and as you suspected Andy that seems to be exactly what the small board is doing.

There are two identical channels, one for each thyristor and then some other stuff down right that I suspect effects the duty cycle or timings. The chip on the right I cannot get an ID on apart from some webpage telling me its a Motorola chip but I'm guessing it helps generate the voltage rails?

I pulled out my old scope and probed around to see if the generated output where at least somewhat balanced like you said and interestingly I found that one channel seems to be correct by generating a small pulse every 50Hz cycle. The other channel is producing 4 pulses per cycle in fairly rapid succession which doesn't seem quite right and matches up to what you were saying about the saturated core and blowing fuses. Do you think this might have blown the diodes as well? I managed to pick up some new ones but I'm holding off on putting them in until I work out what is going on with the thyristors.

I've also attached some more photos of the craftsmanship in the loom in case anyone is interested:)

[oldway]

You should first reverse engineer the power circuit and post the schematic.....
Phase control acts on the primary or in the secondary?

This is important because the behavior of thyristors is different in each of these cases, especially as you write that it sat and hummed.

About the screen capture of your oscilloscope, it seems to me that there is an error: the pulses of the 2 thyristors must be separated from 10 mS and not from 20 mS as shown in the picture.

This is doubtless an error of trigger.
On an analog oscilloscope, it is necessary to trigger on channel 1 or on line and use the chopp mode and not alternate.

[Sr28]

The thyristors are part of a bridge rectifier on the secondary. I've attached drawing of how that first bit goes together. There is a second transformer which powers the other boards and I think also generates the signal for the thyristors. This goes into the smaller of the the 2 circuit boards at the bottom which I called 'yellow' wire (before I realised there was more than one). There is another matching one for the other channel.

My scope is very old and only channel 2 works (project for another day) so I am unsure where the two signals sit in relation two each other but I expect them to be offset by 180 degrees.

[oldway]

It is more logical, it seemed to me that a primary control with unijunctions, this was the squaring of the circle.

Be very careful, this circuit is "hot", very dangerous!

I hope you know the risks of death you run!

The triggering circuit of the thyristors is very simple, it seems to work perfectly.

The load is capacitive, so there should be no triggering problem.

But anyway, Iit would be interesting to see the phase relationship between the triggering pulses and the 50Hz sincronism signal.

NB: don't measure anything on the "hot side" .... look for pulses only at the unijonction and controle side

These 2 thyristors have worked with large dI / dt and may have lost their gate sensitivity.

The trigger pulses of an unijunction transistor is far from ideal, the pulses are very short and of current and voltage are very limited.

[Sr28]

Thanks for confirming that the triggering was correct!

With that in mind I went ahead and replaced the diodes in the rectifier as I knew they where bad. As soon as I did that we now have 711V on the output of the rectifier! Progress!! Those diodes must have been right on the giving that weird 10V output before completely failing short.

The weird thing is that it is now giving about 550V on the output terminals and nothing I do seems to adjust that. The overload light is also on but the unit isn't drawing an excessive amount of current so I expect this is also a fault and it is still defaulting to constant current mode.

I initially expected the output transistors but after pulling one out to test it seemed to be okay. Should I pull them all out to be sure?

Any ideas on if one thing could be causing all these faults? We can't be that far off now surely?

The last thing that I've discovered is that a voltage only displays when the external sense wires are connected. Could this unit not have internal sense wires?

Also any idea on what the Pin 1 (-5V), Pin 2 (0V) mean under the 'Voltage Program' button and likewise on the 'Current Program' button? Are these activation signals or external control signals?

Thanks for all your help so far!

[oldway]

The output transistors are probably conductive because they receive a base current.
You can check this by measuring their voltage between base and emitter ... it will probably be around 0.6V - 0.7V

The thyristorized pre-regulator and the output transistors are thus in full conduction.

The voltage regulation obviously does not work.

Normally this can not be due to a lack of voltage sense information because these power supplies with such dangerous voltages are designed to have no output voltage in case of loss of voltage sense feedback.

This design is imposed by safety rules.

If the voltage goes to maximum, uncontrolled, it is better to look for a fault in the voltage regulating circuits.

But without diagrams, it's hard to help you on this.