|
Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders. |
|
Thread Tools |
19th Mar 2018, 9:10 pm | #1 |
Dekatron
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,517
|
Thorn Automation 40v 50A PSU fault
Me again
I am struggling to get my head around this power supply - it is a thyristor controlled pre regulator, and almost conventional post regulator design, using lots of 2N3772 as the pass elements. From my simplistic view of things, the Thyristors are making up part of a conventional bridge rectifier, the control circuit should drive these harder into conduction as the requested voltage rises - from the parts of the manual I have, it would seem that this is the case - I guess the trigger pulses become more frequent as the call for output goes up. In my case, I have the regulator controls wound up for maximum smoke, but can only achieve a paltry 35V. The transformer is supplying 52V AC to the rectifier pack, and the peak I am seeing out of the pack is the above mentioned 35V. I would like some method of checking that the rectifier pack is working properly, and able to achieve full output, so that I can decide where to look next for any fault, but am having problems deciding the best way to break into the circuit to seperate the link between pre and post regulator (they are both discrete circuit boards. Anyone fancy a stab at guiding me through this over complicated 52KG beast? Cheers Sean
__________________
Engineers make things work and have spare bits when finished |
19th Mar 2018, 10:04 pm | #2 |
Heptode
Join Date: Nov 2009
Location: London, Ontario, Canada
Posts: 672
|
Re: Thorn Automation 40v 50A PSU fault
Hi Sean,
The frequency of the trigger pulses to the pre-regulator thyristor circuit does not change, just the phase of it. Similar to a light dimmer. Check with a scope the change of the trigger pulse phase when the scope is triggered by the AC voltage before the rectifier-SCR stage. You should see the phase moving back and forth as you change the output voltage. A link to the schematics would be helpful! The late Jim Williams of Linear has an application note which, among others, describes a linear PS with an SCR-based pre-regulator, for high efficiency: https://cds.linear.com/docs/en/appli...note/an32f.pdf Regards, Peter |
19th Mar 2018, 10:17 pm | #3 |
Dekatron
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,517
|
Re: Thorn Automation 40v 50A PSU fault
Hi Peter,
I will pop what I have on Dropbox or Flikr (they are too big to put on here) Cheers Sean
__________________
Engineers make things work and have spare bits when finished |
19th Mar 2018, 11:17 pm | #4 |
Dekatron
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,517
|
Re: Thorn Automation 40v 50A PSU fault
__________________
Engineers make things work and have spare bits when finished |
20th Mar 2018, 3:52 pm | #5 |
Heptode
Join Date: Nov 2009
Location: London, Ontario, Canada
Posts: 672
|
Re: Thorn Automation 40v 50A PSU fault
Hi Sean,
This is a really nice PS! I have downloaded the pictures - still piecing the document together, renaming the pages and stitching the schematics together for better comprehension. One more scan would be helpful, the schematics of the "A2-05141-Series Regulator for Mod1" is missing a section from the middle where the page was folded out. Send a PM with your e-mail address, it looks like it will not be that simple, it might take a while, if you are serious about it. Hopefully you have a scope too! Regards, Peter |
20th Mar 2018, 6:37 pm | #6 |
Nonode
Join Date: Jan 2009
Location: Bristol, UK.
Posts: 2,059
|
Re: Thorn Automation 40v 50A PSU fault
A scope accross the main smoothing can should reveal if both SCR's are firing but you will need some load on the PSU to make the voltage 'droop' a bit between the charging pulses. Its possible the SCR's may not fire for every cycle under a light load - not looked in detail at the control circuit. A car headlight bulb should be enough.
dc |
20th Mar 2018, 7:33 pm | #7 |
Dekatron
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,517
|
Re: Thorn Automation 40v 50A PSU fault
Hi Peter,
What you have, is what I have - I obtained that set of images some years ago. Worst case I will have to remove the regulator card from the supply, and draw out the missing section. I do have a scope, so no problem there. Dave, Ok, good idea - will have to load the caps down a bit and have a probe around. Cheers Sean
__________________
Engineers make things work and have spare bits when finished |
20th Mar 2018, 9:01 pm | #8 |
Heptode
Join Date: Nov 2009
Location: London, Ontario, Canada
Posts: 672
|
Re: Thorn Automation 40v 50A PSU fault
Hi Sean,
I hoped you had the complete manual and it was you who scanned the pages! Looking a bit further, pages 28 and 29 are missing too, those contain the circuit description of the Firing Pulse Generator circuit. Oh well, if the description is not available, simulation is always great tool to help understanding the workings of a circuit (or crucial parts of it). As mentioned you could check the firing pulses. It is originating from the B2 or B3 point on the "A1-0297 PS 5040 Pre Regulator" control board. Put one channel of the scope to the 18V AC input of the pre-reg control board (point B7), trigger the scope from this channel. Put the other scope channel onto the firing pulse. As you change the output voltage the phase shift between the AC input signal and the firing pulse should "walk through" nearly the full part of the half sinusoidal AC input signal. Do you see that working? If so, how big is the range? You can see this signals on the "General Circuit Mod 0 & Mod 1" drawing. Regards, Peter |
20th Mar 2018, 10:15 pm | #9 |
Dekatron
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,517
|
Re: Thorn Automation 40v 50A PSU fault
Peter, A copy of the post regulator circuit - I have found that there wasn't much detail missing.
I will refit the board tomorrow evening, and get a scope hooked up Cheers Sean
__________________
Engineers make things work and have spare bits when finished |
21st Mar 2018, 8:04 pm | #10 |
Dekatron
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,517
|
Re: Thorn Automation 40v 50A PSU fault
Slight setback.....
Due to some clumsy misreading of the 0V line, I now need to replace some diodes on the pre reg board... one of the electrolytic caps has also surrendered in quite an impressive manner. Will be the weekend before I can do any more. Oh well, a bit more work
__________________
Engineers make things work and have spare bits when finished |
23rd Mar 2018, 5:09 pm | #11 |
Dekatron
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,517
|
Re: Thorn Automation 40v 50A PSU fault
Scope clumsiness has been rectified (literally - 4 new rectifier diodes and a couple of new caps)
Back to the original fault now. Peter, I have scoped the points you as for - as expected, on the 18v connection, we see a nice sine wave, but connecting to either B2 or B3, we see absolutely nothing! So, I think this could well be an area of investigation - I think what is happening is that the two conventional rectifier diodes are supplying enough DC to the smoothing caps to allow the post regulator circuit to function, but the pre reg is not asking the thyristors to fire at all, hence the rather pathetic attempt at output.
__________________
Engineers make things work and have spare bits when finished |
24th Mar 2018, 3:28 pm | #12 |
Dekatron
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,517
|
Re: Thorn Automation 40v 50A PSU fault
Had a bit more time to play today.
I am now convinced the fault is in the pre regulator circuit - I disconnect the DC supply from the rectifier pack, and substituted a DC feed of 50v from anther supply. The series regulator was then able to control this output voltage from 0-40v when under a 2A load (all my other bench supply is capable of). The fault protection circuit is also certainly working - this circuit triggers a shunt trip breaker on the power supply if there is an over voltage situation, or if the rectifier pack goes over temperature (there is a thermal fuse mounted to the heat sinks that closes on fault - I have bridged across this, and the supply is immediately tripped.) Tricky little beast. So, a question - can I connect an AC supply to the trigger lead of the thyristors to see if I can get them to conduct?
__________________
Engineers make things work and have spare bits when finished |
24th Mar 2018, 3:55 pm | #13 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,801
|
Re: Thorn Automation 40v 50A PSU fault
I wouldn't do that just yet, it too could cause an outbreak of drama.
The drive to the thyristors, as said, resembles a lamp dimmer, running at the mains frequency. the fraction of the correct half-cycle that the thyristor is on is the Oomph factor of the pre-regulator. The pre-regulator's job is to control the voltage on the reservoir capacitor of the main regulator. The load takes whatever current it wants and the main regulator controls the throttle of the hordes of 2N3772s to get the output voltage to whatever the operator has set. The pre-regulator controller looks at the output voltage going to the load, looks at the reservoir capacitor voltage and subtracts them to determine the voltage all those 2N3772s are dropping. You want several volts of drop so the main reg can respond to load changes, but you don't want more than that. The pre-reg controller adjusts the oomph from the thyristors to keep the reservoir x volts higher than the output. This saves heat. So the pre-reg control depends on the main reg. It isn't an independent item. You could bridge the thyristors with normal diodes, making a normal bridge. The reservoir should charge to full welly and the main reg should work normally. It won't have normal endurance if loaded to full current because of the greater heat dissipation in the heatsinks. But you should be able to run and check the main reg before turning to the pre-reg. I've not reassembled all those bits of diagrams, I'm working on memories of a different brand. But I expect the same modus operandi. David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
29th Mar 2018, 3:30 pm | #14 |
Hexode
Join Date: Nov 2003
Location: Ossett, West Yorkshire, UK.
Posts: 482
|
Re: Thorn Automation 40v 50A PSU fault
Don't apply AC direct to the gates, this will almost certainly destroy the SCR's
How easy is it to scope the SCR firing card? Check that terminal B9 is at ground potential, if not temporarily remove the earth from the scope mains plug, my scope is always earth free anyway. Connect the scope signal earth to anywhere from B9 line. If you can sync the scope from the line input, otherwise use channel B connected to the junction of D8 and D9. Set the time base to 2mS / Div. The collector of TR4 should be at about 15 volts most of the time, dropping to zero coincident with the zero crossing point. Next have a look at the collector of TR6 for a ramp of several volts. If that is OK look for 0.7 volt pulses at the base of TR9 then pulses of + and - 15 volts at TR10 emitter. There will probably be two pulses, the first coincident with zero crossing and the position of the second should move when the voltage control is adjusted, if not have a look at the collector of TR12 and see if that changes with the voltage control Last edited by See_Mos; 29th Mar 2018 at 3:46 pm. |
29th Mar 2018, 4:46 pm | #15 |
Hexode
Join Date: Nov 2003
Location: Ossett, West Yorkshire, UK.
Posts: 482
|
Re: Thorn Automation 40v 50A PSU fault
I had another look at the diagrams and the B9 line is connected to the positive of the 15,000uF capacitor so as I suggested remove the scope earth or use an isolation transformer and be careful not to let the scope case touch anything.
|
29th Mar 2018, 5:51 pm | #16 |
Dekatron
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,517
|
Re: Thorn Automation 40v 50A PSU fault
Thanks! - I will take a look at these points later this evening - progress report to follow.
__________________
Engineers make things work and have spare bits when finished |
30th Mar 2018, 9:54 am | #17 |
Hexode
Join Date: Nov 2003
Location: Ossett, West Yorkshire, UK.
Posts: 482
|
Re: Thorn Automation 40v 50A PSU fault
I made a mistake about the waveform at TR10 emitter, the base line will be at -15v with a very narrow vertical pulse
Last edited by See_Mos; 30th Mar 2018 at 10:04 am. |
30th Mar 2018, 5:04 pm | #18 |
Dekatron
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,517
|
Re: Thorn Automation 40v 50A PSU fault
The collector of TR4 should be at about 15 volts most of the time, dropping to zero coincident with the zero crossing point.
The collector is sitting at around +15 volts all the time Next have a look at the collector of TR6 for a ramp of several volts. Looks to be about 6v p-p If that is OK look for 0.7 volt pulses at the base of TR9 then pulses of + and - 15 volts at TR10 emitter. There will probably be two pulses, the first coincident with zero crossing and the position of the second should move when the voltage control is adjusted, if not have a look at the collector of TR12 and see if that changes with the voltage control TR10 Emitter has positive pulses of around 0.5v, with the whole point sitting about 0.5 volts below 0V The pulse duration does change slightly when the voltage control is advanced, and pulses drop off completely when asking for 0v output
__________________
Engineers make things work and have spare bits when finished |
31st Mar 2018, 11:46 am | #19 |
Hexode
Join Date: Nov 2003
Location: Ossett, West Yorkshire, UK.
Posts: 482
|
Re: Thorn Automation 40v 50A PSU fault
Looking at it again it really is a strange circuit and I am probably wrong about the pulse on TR4 collector. It would be so much easier with the pre-regulator board in front of me.
I wouldn't expect the pulse to disappear completely as you still need a few volts across the series regulator even for a low output. When set for maximum voltage, relative to the half cycles on CH B, how far back does the pulse move? The signal at TR10 E is smaller than I would expect, is TR10 OK, an open collector could explain this. Also if C9 is a 'lytic and it is faulty there might not be enough energy to fire the SCR's fully, I would replace it it with a 2u2. Last edited by See_Mos; 31st Mar 2018 at 11:59 am. |
31st Mar 2018, 10:16 pm | #20 |
Dekatron
Join Date: Feb 2005
Location: St.Ippolyts, Hitchin, Hertfordshire QRA IO91UW
Posts: 3,517
|
Re: Thorn Automation 40v 50A PSU fault
This really is a difficult thing to work around.
The pulse does not move in relation to the AC cycle at all. I am wondering if there are more areas of the circuit interacting here. The Input labelled P comes from the protection board - the description of operation leads me to understand that this board is only fitted to provide over voltage protection - if the voltage threshold is exceeded, the protection board removes drive to the pre reg board, and tires the shunt trip on the mains inlet breaker, thus shutting the PSU down completely. I wonder if this input to the pre reg board is sitting at a level that is shutting the firing pulses down? What I would really like to do is test each module individually, just to prove that any particular fault is located in one section only - at present, everything seems to interact.
__________________
Engineers make things work and have spare bits when finished |