Help needed TTI TSX3510A 35V 10A PSU error E118

[chriswood1900]

I need some help fixing a TTI TSX3510A 35V 10A PSU error E118
These TTi PSUs have a mix of switch mode and linear, the initial conversion from mains to DC is done by the switch mode part which provide DC at approx. 1.5V above the linear part output stage, this limits dissipation at high current levels.

The fault is when you switch the output on it immediately trips with error E118 Output stage has tripped (OVP or Temperature)
When looking at the output it zooms upwards until the control system turns it off.
On the bright side all the programmable part seems to work, you can programme voltages and currents and all is OK until you press the output button.
Does anyone have experience of repairing these and know any common faults which cause these conditions. Mine is an issue 5 from 1996
A manual is attached for those that might be interested

[G6ONEDave]

Perhaps a shorted output transistor Q9 to Q12 or their regulation control section, MJE3055 etc.
Dave

[dave cox]

Are they driving HC logic from NMOS ?

[dave cox]

It looks like D11 should be conducting (into VCON) as the output voltage heads towards the desired OP voltage, shunting Q8's base current, limiting voltage across R36. They have put the load into the collectors of the series pass transistors ...

Whatever is driving VCON is mis-programmed or it's broken (assuming you mean it zooms PAST where it is supposed to stop).

I hated every second of fixing my TTI unit, packed, fragile PCB, horrid DIL sockets and dodgy TTL no to mention the mechanical layout and access!

dc

[chriswood1900]

Dave in the 3510 the two output transistors are Q14 and 15 and they seem to be OK, the manual covers the 1820 as well and you have to work out which bits are in.
dc I have just taken a few basic measurements so far checking the auxiliary supplies +/-5 and 12v are ok which they are. Over the weekend I will try tracing VCON and yes it zooms past where it is supposed to be and shuts down. I have not looked beyond the main board yet.
Chris

[Chris55000]

Hi!

A manual is attached for those that might be interested

Hi!

I am and will come forth with fault–finding advice when I've studied it!

Chris Williams

[chriswood1900]

Thanks for the replies so far I have managed to find time for a little further testing.
In the manual P17 it sets out a procedure for testing the linear output stage by first disabling the SWPS section with PJ4 and powering it from a 5V PSU. You then test out the linear section and set the pre-set controls.
I was unable to set VR3 for zero output there always being around -0.5V at the output whatever the setting. Moving on a stage further I disconnected the 5V supply and the -0.5V remained charging up on switch on and slowly discharging when mains power is removed. The only source of power at this point was the supplies from T1 providing +/- 5V and 12V and the isolated approx. 5V for Q4.
The voltage references in this supply is against the + (plus) output rail so the output has +0.512V against the positive terminal and hence a negative output, I measured the current available and it is approx. 1mA at which point the voltage collapses. I first thought there might be a power leak through Q4, Q29 & Q8 but I can’t measure anything so at the moment I’m a bit stumped.
I'm not sure if the above relates to the initial faults but it seems this may indicate other problems. Has anyone else experience of repairing the output section?

[chriswood1900]

Well I have at last made some solid progress and in the process improved my understanding of these PSUs.
It has to be said these are no the easiest supplies to work on, they can be very dangerous with many items such as heat sinks at HV DC.
They have a double-sided PCB with tracks in all directions and virtually all the component location markings covered by the part, not all parts are adjacent to where you might expect them so there was a lot of time consuming tracing through trying to locate a part and the large number of parallel paths made any in-circuit testing difficult.
Having not found anything that was quick and simple and chasing a non existent fault when I had -0.5 volts on the output this was intentional and is supplied by R80.

Starting again I stripped the unit out of its case and started the manuals fault finding procedure.
First part was to disconnect the Switch mode and just power the low level control circuits from a the transformer T1 and an auxiliary 5V DC supply to the linear output section.
This threw up a problem with the output wandering from minus 0.5 volts to plus 5V and the control VR3 could not set the 0 Volt point.
I checked all the DC supplies were in order and was getting the +/- 2.45V from the reference diodes. The next item I suspected was the LM724 Quad Op-amp (IC5) but that seemed OK and I tried substituting a new one but no change. The output on pin 8 was sensible and varied smoothly as you adjusted VR3 the inputs on pins 2&3 were zero as expected but pin1 which controlled the output was wandering all over the place pretty much randomly. It was now time for a quick crash course on Op-amps to understand better how they worked to carry on fault finding. The job was picked up and put down multiple times as normal work and other things arose. I had a conversation with another forum member and came to the conclusion that there must be something on the feedback circuit causing the problems. I decided to lift the 2 feedback resistors R64 and R68 220K and this is where I made the discovery when checking the resistances that R64 was intermittently open circuit but showing no other sign of distress.
I ordered new 220K 1% resistors from RS along with some capacitors that I decided probably needed changing. I normally check the ESR on caps and leave them if they test OK and are a quality brand, in this case Elna. They all tested OK but I replace the ones that operate at high frequency in the SMPS section as a matter of course.
New parts fitted I went through the calibration procedure and all tested good and I now have a 35V 10A PSU which hopefully is reliable.
A few learning points on these, follow the manual that I posted earlier and break down the various sections. It is important that the linear output works properly or you can’t test the Switch mode part as there are so many fault detection circuits it stays shut down. Almost all voltages in the linear part reference from the positive terminal contrary to my normal expectation.
I hope this proves useful to others who may be tempted to tackle one of these.