Farnell L30/5

[Diabolical Artificer]

Have one of these on the bench given me some time ago by a kind forum member. It had no OP and the ammeter was shooting to FSD and back to 0. Checked voltages as per schematic and though low, most checked out. Checked all pass trannys and driver etc.

Eventually found R104 OC, replaced, now have OP. However it only goes from 5v to 300v
and the current limit isn't working. With a 10 ohm load the ammeter goes to FSD.

I'm not 100% certain I have the front terminals connected correctly, not having used one before. I have the +/+ sense and -/-sense terminals bridged.

I checked the voltages on IC2, these are pins 1) 1.4v 2) 0v 3)0v 4) -5v 5) 0.6v 6) 0.5v 7) 12.5v and lastly 8) 14.5v (15v). AFAIK the RHS part of IC2 is the current control, the left controlling voltage.

A few questions apart from the obvious, what is the purpose SW103? What does "set C/L" mean, set current?

TFL, Andy.

[TonyDuell]

I would guess that 'set C/L' is 'set current limit'

[MrBungle]

Set C/L is set current limit.

SW103 sets the entire supply into constant current or constant voltage mode.

If it's going wonky I would check R103 and SW103 actually are actually actually still resistors and switches

Also check reference voltage at TP20. Looks like everything is derived from that. Nothing will work properly if that's not right.

[mhennessy]

The full manual can be found here: https://www.vintage-radio.net/forum/...08&postcount=2 - many of your questions are answered in there

SW103 is to put the unit into a more precise current limit mode than usual. Can be left in the CV position ordinarily - the unit will still enter constant-current mode when the limit (set by the "C/L" control on the front panel).

I assume that "300v" is a typo - unless something is spectacularly wrong, the highest voltage you'd get from one of these under a more sane fault condition would be around 50V. Working, the output should vary between 0 and 30V, of course.

Are you sure you've used the correct 0.1 ohm resistor for R104, and are there any broken connections or dry joints in that area? The current meter relies on R104 as its shunt, so if it's going FSD with a relatively modest load, that does point in that direction.

Does LED102 light up when the unit is loaded?

[cmjones01]

I have one of these which does odd things when it hasn't been used for a while. The ammeter will shoot over to its end stop, for example. It's due to the Molex connectors on to the PCB inside not making good contact. I take the lid off and wiggle them and it's OK for another year or so.

Chris

[Refugee]

I would first go for the negative reference voltage for the sensing circuits.
The fact that it fails to go below 5 volts is the clue. Check the voltage across Z2.
The current metering may well settle down once the negative rail is correct as it is used by all three chips.
I have no idea how much of the circuit will continue to work with a blown F1.
Check it anyway.

[mhennessy]

I think the aux rails (hence F1) are OK because there's -5V at pin 4 of IC2, and +14.5V (close enough) at pin 8.

The current meter is in parallel with the shunt resistor, so doesn't rely on the ICs to give sensible readings...

Pretty sure that bad connections will be part of the problem.

[Diabolical Artificer]

Thanks, C/L = Current limit. TP20 was 6.1v and all voltages as given on schematic are about right. I measured from the ground end of R7 instead of pin 7, but they're essentially the same thing, IE ground/0v. My results were low, but I measuerd them with a Fluke DMM 1M not 20k/V.

Will check SW103 etc. As I said I havn't used a current limited PSU before and wondered if I was using it correctly. I also don't know what it should do if working correctly. So should this PSU work only as a CCS or CVS?

Yes 300v is a typo Mark and ta for the link.Yes 100% sure I fitted R1 resistor. I have quite a few 10w ali jobbies after I mis ordered for 1 ohm - 1R when I first started electronics. I also checked it on several meter's inc the excellent one Mr Bungle sold me.

When I checked the original R104, several chunks of the coating has blown off, so a severe over current condition has occured in the past. I checked the off board components around R104, EG D102/103 which I presume are protection diodes. D102 did behave oddly though, which at the time I put down to my DMM not having enough test voltage or something.

No LED102 doesn't light, but VT4 is ok, will check.

First thing I checked mate, the fuse : ) It's fine.


"Pretty sure that bad connections will be part of the problem." I had one connector off, the pins original coating had tarnished. I gave it a zizz with a brass wire brush but will redo both and report back.

Muchus gracias amigo's, Andy.

[Diabolical Artificer]

Progress of sorts, after cleaning and abrading the pins and pulling the pins in and out of the sockets, V out and the voltmeter goes down to zero.

However with a 10 ohm load, the ammeter tracks the voltage, EG 10v = 1A, 20v = 2A and 30v = 3A. This is with the current pot at "zero"

With the voltage pot at,zero OP, load switch up, I'm getting some odd readings. At the junction of D2 and D3 I'm getting 0.75v (2.8v) and at pin 10 I'm getting 0.06v (0.6v). Now on the schematic it says to take the reading at 30v 5A! The biggest 6 ohm resistor I have and jump leads are frying at that and the OP fuse blows. I did take readings with the OP at 30v, load switch up and got 1.7v and 0.08v respectively.

I rechecked all trannys on the board and all check out ok. I checked all diodes and resistors too and replaced the IC2's socket and cleaned IC2's pins. I also checked SW103 and R103. SW103 checks out, R103 was low and dis-coloured so I replaced it. Lastly I found a RIFA cap and removed it - C103. I'll replace it with a Y rated cap when I get one.

So scratching my head here. Could IC2 be duff? I checked it and IC1 for shorts, they're ok there. If someone could run me through the circuit it may help.

AFAIK VT1 and VT2 and associated components form a voltage reference. With IC1 and part of IC2 etc these are a variable voltage source. The other part of IC2 should control current. FBis fed back from the sense resistor to IC2 pins 5,6,7 which is a comparator. This should control VT3, which should rob current from VT101's base, thus controlling current OP. Is that roughly correct?

Oooh, one last thing LED102 was SC. Someone had replaced it in the past using a lot of flux as the leads were corroded. This LED just fell off the flying leads. Oh and there's a diode not marked on the schematic D4 on the board. It is connected between pins 7 and 13.

Think that's it. Will download manual and read.

Cheers, A.

[mhennessy]

Yes, before doing anything else, check that SW103 is set to "CV". And treat it to a shot of switch cleaner and some exercise. Also, make sure the front panel "Current Limit" control (P103) isn't turned fully anti-clockwise. When at minimum, all sorts of strange things might happen

On the assumption SW103 already is in the right position...

The "LED not lighting" symptom is interesting.

In "normal" (CV) constant voltage operation, IC2a controls the output and IC2b has its output (pin 7) rammed up against the positive supply rail. But once the output current has approached the limit set by P103, then the output of IC2b falls down away from the rail, and takes control of the output via D3. VT4 is arranged to detect this happening, and light the LED. If you look, VT4 is conducting when in CV mode, and shorts out the LED. VT4 stops conducting when IC2b's output moves down away from the 15V rail.

So is VT4 correctly reporting that pin 7 is not pulling down the output voltage via D3, or is VT4 faulty? The easiest way to test VT4 is to short between base and emitter. That should prompt the LED to light. If not, investigate VT4 and the surrounding resistors, connectors and possibly even the LED itself... But if all that proves that VT4 is indeed doing the right thing, then we know that we need to focus on IC2b and its surroundings... But check SW103 first!

Regarding the diodes D102/103, yes, they are protection diodes, and are reverse-biased in normal operation. When you say D102 behaved oddly, this needs further investigation. It's a fairly regular diode that has a 0.6V PN junction, so a DMM will have no problem with it. For testing, unsolder one end and see if that improves things. If it's leaking, it might be why the unit won't go below 5V with no load connected (assuming SW103 is in the right position). If it proves to be faulty, just about any rectifier diode would replace it - even a 1N4002 or similar, as it won't ever pass significant current (unlike D103, which does need to have a reasonable current rating).



A brief (!) overview of using power supplies like this:

All bench power supplies are current limited, but what makes this one different is that you choose the value at which the unit enters current limiting - hence the Current Limit control on the front panel.

Let's say the control is at max (fully clockwise). In this case, the current limit will be 5A (slightly over that in practice). Providing you are drawing less than 5A from the supply, then it will behave as a decent voltage source - the output voltage will stay resolutely at the value you set with the voltage controls (CV mode). But when you reach the current limit of 5A, the unit enters CC (constant current) mode, and whatever happens, it won't let more than 5A flow - even if you put a dead short across the output terminals. Naturally, as you reduce the current drawn, the unit will seamlessly go back to CV mode.

In practice, you'd set the current limit to a much lower value for typical bench work. Let's say you're working on a transistor radio. In this case, I'd start with a current limit of around 50mA, and if the radio is OK, then it won't draw as much current as that with the volume kept nice and low, so the PSU will stay in CV mode, delivering a steady 9V, say. But should there be a problem with the output stage, for example, then the PSU will simply enter CC mode, and no more than 50mA will flow - and this action could well save the life of the output transistors, or prevent a leaky decoupling capacitor from overheating and exploding.

These Farnell units do have one problem: how do you know what the current limit actually is? You've only got an approximate idea from how far you've rotated the Current Limit knob. If you need to know more precisely before you connect up the load, then shorting the output terminals while watching the current meter is how you do it. Fine for low currents, but if you're trying to adjust to a higher limit of several amps, then do it quickly! A short-circuit across the output of a linear series-regulated supply like this represents the worst-case condition for heat dissipation in the power transistors. Luckily, this unit has plenty of them, and decent heat sinks

The Thurlby Thandar units (PL320, etc) provide a preview of the pre-set current limit when the DC output switch is set to "off". This is very handy - also, it's very easy to do electrically, so it's a shame Farnell didn't do it.

Anyway, hope all this helps - this will be a nice addition to your workshop once sorted, and it's definitely worth spending a bit of time thinking about how it works. I built several power supplies like this years back, and I recommend that anyone interested in electronics should do the same - it might not be an audio amplifier, but it's a good way to learn about analogue and power electronics - all of which builds knowledge and experience that will feed back into your audio work

[mhennessy]

Oops - posts crossed

I won't reply in detail to #9 at this stage, other than to say that based on the comment about the current pot at "zero", I'll let you go away and read post #10 and the manual before commenting further.

[mhennessy]

Hi Andy,

Another quick one. If you'd like to understand the circuit a bit better (though I'd say you're not far off, honestly), then take a look at this PDF - the explanations start on page 19. This is almost identical to the Farnell unit - it's a project people build (and keep) when they attend a course I run at work.

https://www.dropbox.com/s/cc0t1ozuj0...k%202.pdf?dl=0

This is the schematic:

https://www.dropbox.com/s/jjaymhqzvw24lcc/PSU.gif?dl=0

Hope this helps,

Mark

[Diabolical Artificer]

Thanks Marc, I'll go and do some reading and do as you suggest.

Get what you mean about it not being an amp but useful to research anyway. I learn more by fixing and sussing out circuits like this than complicated amps; there's too much going on and it's harder to read the circuit.

A.

[Diabolical Artificer]

Followed Marks advice and shorted VT4's B&E, LED102 lit up. Next looked at IC2b's OP. At low voltage OP it stuck at V+ as Mark predicted. Cranked the voltage up, it stayed there, therefore IC suspect. Tried a few generic opamps, IE JRC4565, aha! LED102 lit up briefly then, opamp died. Put in a TL072, success! PSU went into CCS mode.

I'll get a proper replacement but for now very happy this got fixed. Thanks to you all, big up and good vibes going out to Mark.

Andy.