Xenoid - 24-2-2013 at 17:30
A few years ago, I purchased one of these supplies (LKB 2103), it's about 30 years old. I used it initially for a few experiments, but then left it
for a couple of years.
When I turned it on the other day smoke issued forth! Upon examination the conflagration proved to be centered on a small inductor marked L501 which
was totally destroyed.
The circuitry in these supplies is extremely hard to trace and is made doubly so by the fact that there is no manual or schematic available online.
Does anyone, perchance, have one of these supplies or access to a manual/schematic!
At the very least can someone tell me the value of L501 (it's on the main board, between the two plug-in boards, and adjacent to a largish capacitor.
I soldered in a 270uH inductor so I could check the current it was over 2 Amps, judging from the size of the inductor wire (~0.07mm) it should be only
a few mA. The problem seems to be on the front plug-in board but I haven't found any faulty components yet.
I live in hope..... 
IrC - 24-2-2013 at 21:29
Most likely a high voltage electrolytic went bad, leaky or whatever. The current is coming from high voltage rectifier diodes which shorted from the
load of a bad capacitor. You need to replace any bad capacitors and diodes. Odds are the choke was in the millihenry range if it is simply a filter
choke but I doubt it is that critical. I cannot imagine why they would need to block RF in a simple DC supply for the designed purpose. You never know
I suppose. Is it a ferrite or steel (laminated) core or air core? Knowing this, the size of the original choke, the type core, and expected current in
normal operation would allow you to make a fair guess. You do not need a schematic to repair it. Just some idea, attention to detail, and so on. If
you have not already damaged the HV transformer you are surely risking it by turning it on while any bad diodes and/or capacitors are still connected.
Your first test should be to isolate the secondary, measure the AC voltage with proper meter, and run it for a little while to be sure the transformer
is not heating unusually. What voltage and current is it supposed to supply?
I should add if a capacitor is leaky, it may be hard to test at low voltage. Meaning if you replace the diodes and turn it back on, the leaky
capacitor may short a diode again the second high voltage is applied to the capacitor(s). What are the capacitors?
[Edited on 2-25-2013 by IrC]
Xenoid - 24-2-2013 at 22:16
Thanks for the reply IrC
Unfortunately, this is no ordinary simple power supply, if it was, I would have fixed it by now
The LKB 2103 is a high stability, constant voltage, current and power. There are no less than 6 ferrite "pot core" inverters scattered around. Whilst
there is a mains transformer/rectifier/filter capacitor "front end", it is only to provide low voltage DC for the inverters.
Yes, the inductor is a tiny component, there are a few 22uH and 150uH inductors nearby. I assume they are to block some of the "hash" from the
inverters. The burned-out one looks similar so it's in the same range, I calculated the inductance as about 8uH for a "free air" coil, but it's got a
ferrite core so the true value will be somewhat higher.
The inductor is not really the problem, it's just a victim of too much current.
I really do need a schematic, the circuit boards are hand designed, the V+ and V- are not marked and wander all over, their location only being marked
by the occasional presence of TAG capacitors.
IrC - 24-2-2013 at 23:01
Being a switching supply then chokes in the uH range to block RF make sense. Still I doubt the value is overly critical as long as the wire gauge is
large enough for the current. Maybe around 270 uH 1 amp or thereabouts for the choke. Sounds like a real nightmare repair. But I have done jobs
without any information as hard, go through and check all high voltage diodes and replace any you find bad. You have it narrowed down already to only
one of the switching supplies, the one with the burned choke. You never said the ratings on the capacitors. I do not think it would be impossible to
test every part in the circuit which failed. I still think with enough patience you can do the job even if you cannot find a schematic. Thinking too
much about it can make the job harder. You end up overwhelmed looking at it all instead of focusing in on only the failed circuit. Isolate the failed
circuit and see if the rest of it operates properly.
You said "V+ and V- are not marked". I assume this is the low voltage inverter supply? If this is working I do not see the need to trace it. Focus on
the potcore fed by the choke. Trace it's secondary to the diodes and filter. Test it all. Test whatever switching devices (mosfet or bipolar?) for
shorts. Did one just fail, did it fail from a short in the HV section. I still think if you focus only on this one area it may be a simpler repair. I
do not envy you however.
[Edited on 2-25-2013 by IrC]
jock88 - 25-2-2013 at 16:36
There appears to be a few for sale in ebay (if thats any addition?)
http://www.ebay.com/itm/LKB-Biochrom-2103-Power-Supply-Power...
http://www.ebay.com/itm/LKB-Biochrom-2103-High-Voltage-Elect...
[Edited on 26-2-2013 by jock88]