Farnell TOPS 3D Power Supply - looking for service manual

[andrewferguson]

This was a skip find from last week. It powers on and the 4v - 6v section appears to work OK, but there is an issue with the 5v -17v section. The potentiometer to control the voltage spins freely (I can keep turning it round and round), and the voltage outputted and displayed is stuck at around 2v (considerably below the minimum of 5v).

I'm hoping it's just a matter of a replacement potentiometer and some calibration, but in order to determine if the potentiometer is at fault I need to get to it.... which so far is proving easier said than done. I've got the top/side panel off, and the bottom panel off, and I have undone the screws that I thought were holding the face plate on, but it won't budge. None of the front panel controls seem to be connected to PCBs inside the unit (so I should be able to remove the front panel with the knobs still attached). Indeed, I am able to wiggle the right hand side a bit, but the left hand side still seems firmly attached and I can't figure out why.

Does anyone have the service manual for this so I can try to figure out how to get it apart? (I've found the schematic, but that isn't helping with the disassembly!)

[FIXITNOW]

the circuit is here
https://www.eserviceinfo.com/downloa...%20Supply.html

but no disassemble info sadly

[Julesomega]

Farnell manuals never give any help on such mundane matters as access. The variable for the main output is a ten turn type. The page on "Maintenance" says - you guessed - return to Farnell

[andrewferguson]

A brief update (and a plea for help!) on this.

After a bit of investigating, it seems that the only things holding on the front panel (after removing the screws) are the two knobs to adjust the two voltages. After a bit of experimentation I've discovered that the cap of each knob comes off, and reveals.... something (see attached picture) .... underneath that I suspect is the key to removing the knob.

I've never seen this before, and am not sure what sort of tool would be needed to remove the knob. I've tried a flat-head screwdriver but none of mine fit. Does anyone have any ideas?

[wd40addict]

It's a collet knob. Try a pair of small pliers, one jaw tip in each slot and turn anticlockwise.

[Silicon]

The tool to remove the nut looks like a flat bladed screwdriver with metal removed from the centre of the blade.

I have modified inexpensive screwdrivers to fit unusual applications such as this.

[G6fylneil]

Often the nuts on collet knobs protrude past the end of the inner part and can be undone with a normal screwdriver.

[Herald1360]

Originally Posted by Silicon

The tool to remove the nut looks like a flat bladed screwdriver with metal removed from the centre of the blade.

I have modified inexpensive screwdrivers to fit unusual applications such as this.

Or more likely a tube with appropriate size internal and external diameters with a couple of protruding pegs on the end to engage with the collet nut.

Pointy noise pliers are usually OK, just don't lean too hard using a nice pair of Lindström ones

[andrewferguson]

A long-overdue update on this. I was finally able to get the knobs off this afternoon, using a very improvised solution involving a pair of scissors and two adjustable spanners.

Unfortunately it seems that luck was not on my side... because after all that, the potentiometer isn't actually faulty! If I had bothered to remove the display / multimeter board, and checked the terminals of the potentiometer, I would have saved myself quite a bit of effort.

So, here are the symptoms. The 4v to 6v section works fine. The 5v to 17v, and the -5v to -17v sections both have faults. The 5v to 17v section has a steady output of 2.64v, regardless of the position of the potentiometer. The -5v to -17v section is stuck at 0v.

I'm honestly not concerned about the -5v to -17v section, a negative rail is not something I will ever likely need (and I have a separate power supply with a negative rail if I ever do). I'm only concerned if it is interfering with the 5v to 17v section (ie: a fault on the -5v to -17v section is causing an issue on the 5v to 17v section).

I'm not really very experienced with this sort of repair, so would really appreciate guidance as to how best to track down this fault. I have attached the full schematic to this post. Realistically, if the fault is complex I suspect it will be beyond my abilities to diagnose (actually replacing components shouldn't be too difficult, as it's just a simple through-hole PCB). But I can give it a shot!

[bill knox]

Hi Andrew

You don't say if you have checked any voltages while you have had this unit, there are 4 fuses and it would be useful to make a start and check those , make sure that you have unplugged the unit from the mains supply and then check the fuses, after doing that check that you have the 28Volts at points25 -26 and also at points 27 - 28, then check that you have 28 volts at the collector of the 2 X 2N3055 transistors, These maybe mounted on the back of the PSU and may have protective covers on to prevent S/Cs if anything touches them and other parts of the circuitry, if there are covers on them, remove them for the purposes of this test but put them back on when you have done the test, one thing I can tell you is that of the 4* 2n3055, 2 of them are for the 6-8 volts and of each of the remaining transistors one is for the Neg volts and one is for the Positive volts.

The other thing I forgot earlier is regarding the 4 fuses, I don't know if they are mounted on the back of the PSU or they are inside the PSU.

The other thing that you do not say is, are there any segments of the display alight or is it completely blank.

Regards

Bill

[andrewferguson]

Hello Bill,

Many thanks for your help (and sorry for taking so long to actually get round to testing these things!).

The fuses are a good point - when I pulled the PSU out of the skip, I noticed that both the main fuse and the fuse for the -17V system were blown. I replaced both, then promptly forgot all about them since I got completed distracted by thinking that the potentiometer was at fault (as per my earlier posts!).

I checked the fuses again and, perhaps not unsurprisingly, the -17V one had blown again. So that will be why the -17V output is stuck at 0V (although of course there will be an underlying fault that is causing the fuse to blow).

I have checked the test points you mentioned. The collector for the 2N3055 for the 17V output is at 32.4V (so a little above the 28V indicated on the schematic). Unsurprisingly, the collector for the 2N3055 for the -17V output is at 0V, since the fuse is blown!

Pins 27 and 28 on the transformer give 24.6V AC, as do pins 25 and 26. (Interestingly, it seems that the wiring from the transformer to the PCB is different to the PDF schematic: the schematic has transformer pins 27 and 28 going to the 17V output and 25 and 26 for the -17V output, while in my PSU these are swapped. Not that it is particularly important!)

Regarding the display - it seems to work OK. When on the 4V-6V setting, it accurately measures the output voltage (I haven't checked the ammeter function yet). I have, however, disconnected the entire display board from the PSU to allow me to access the various test points on the main PCB. The 4-6V output still seems to work fine, so I hope that this is an OK thing to do?

Many thanks for your help!

Andrew

[frsimen]

Hello Andrew,

The +17V output of the power supply is designed to track the output of the -17V supply. If the negative supply is missing, the positive supply won't work properly either.

There aren't many things to check to find the cause of a blown fuse in this circuit. I expect you will find a short circuit somewhere. There are only two or maybe three components which have any chance of passing sufficient current to blow the fuse. These are:

C4, D4 and much less likely, IC2.

Also have a look for any wire whiskers or solder blobs that may be causing a short circuit and check that VT4 is insulated from the heatsink if it is supposed to be. I don't think the fuse will blow even if that is wrong, however.

Paula

[andrewferguson]

Hello Paula,

Thanks very much for the suggestions. I have checked C4, D4 and the power rails of IC2, and there were no short circuits. I checked C4 and D4 by putting my multimeter into continuity test mode, and checking the test points 29 and 30. There was no continuity, and the resistance was around 1k. I also checked test points 32 and 33 (the equivalent for the +17V rail) and there was also no continuity with a resistance of around 1k, so it looks like that is how things should be.

I wasn't 100% sure what you meant by "check that VT4 is insulated from the heatsink" - I ran a continuity test between the metal casing of VT4 to the large heatsink it is attached to, and there was no continuity. Is that what you meant?

So all of those things came up blank, and I'm not at a bit of a loss as to what the issue is.

The only thing I thought I'd mention is that during the above testing, I noticed that the switch SW3 (for the -17V supply) had failed in an open-circuit condition. The schematic is wired in such a way that test points 37 and 29 should always be connected, but the switch had failed in a condition where they were not. I don't suppose powering the PSU on like this would have caused the fuse to blow? (In all the measurements today, I had bridged both that switch and the fuse so that everything was in a "normal" condition, but when I initially powered on the PSU, the switch would have been in this failed state).

Many thanks,
Andrew

P.S. When debugging, I kept getting annoyed that the schematic was split across two pages. Rather than the old-fashioned trick of printing it out and taping the two pages together, I have combined the two electronically (attached). This should hopefully make quick inspection of the schematic easier.

[frsimen]

Hello Andrew,

The stitched together circuit is much easier to follow.

For the fuse to blow, an excessive current must be flowing somewhere. If all is working properly, the current limit circuit should prevent the current from getting high enough to blow the fuse. If something were to go wrong with the regulator circuit, for example a collector to emitter short on VT4, the voltage applied to the output terminals would be close to the 32V which is present across C1. The voltages on the circuit were measured with the power supply loaded, which is why your measured voltages are higher. C4 is only rated at 25V and will probably break down with 32V applied, drawing a lot of current and blowing the fuse. It may still measure normally when you test it at a the usual few volts used by capacitance meters etc.

I suggest that checking transistors in the regulator circuit, in particular VT2 and VT4 for a collector - emitter short. You could try running the supply with C4 disconnected, or replaced with one capable of withstanding at least 40V,. This should stop the fuse blowing and allow further fault finding if you don't find anything wrong with VT2 or VT4.

Edit: If you do this, C9 could be damaged should the -17V go beyond -25V, as the +17volt supply will try to track the -17 volt supply. It would be best to disconnect C9 or replace it with a 40V or higher rated component.

With the fuse blown, you will not measure anything useful from the negative output terminal, you need to use pin 7 on the PCB or the negative terminal of C1 as your reference point.

The switch fault you mentioned is unlikely to lead to the fuse blowing.

Paula

[Dekatron]

I Googled some and from what other people write on repairs of this and other similar models from Farnell they usually fail with some or all of the transistors (BC182PL, 2N3053 & 2N3055) and sometimes even with the 741's being damaged.

I've had similar problems with similar power supplies from other manufacturers and the easiest repair has always been to replace the transistors and the 741's to get them up and running and while I've been repairing them I've replaced all of the electrolytic capacitors just as "frsimen" mentions. In rare circumstances the rectifying bridge has been damaged.

Sometimes I like a challenge when repairing something but usually I replace semiconductors as they are often the culprit when a power supply has stopped working.