While it may not fall into my current theme of ‘Items committing suicide when in close proximity to me’, this vintage power supply (from the early 1960s) was lucky enough to survive a sudden attack of conscience on my part, after I chose to replace the missing grommet from its mains cable after I decided to install it at the back of my bench, instead of storing it between uses.
As an aside, I was surprised to find an identical L30/T currently on sale from the US for £23, but being heavy, costing a further £85 for delivery. Operating voltage was the standard 210 – 240 V, set by a voltage selector on the rear. They did make 105 – 120 V models, but despite being in the US, the one for sale wasn’t one of those.
This one has apparently never had a grommet, maybe not even when new, as there’s no evidence of discolouration around the cable hole, or marking/indentation on the lead itself. There’s never been an issue since the mains cable usually spent its life coiled up on the rear of the supply, but its new home means the full length of the (short) mains lead was probably going to be needed, so I thought I should make this ‘right’.
This dual supply is basically two identical, but separate/isolated, 30 V 500 mA DC supplies with current limiting, mounted within a single enclosure. Single versions were also sold.
Voltages can be set with a fair degree of precision – a switch selects one of three ranges: 0 – 10; 10 – 20; 20 – 30 V, while a variable control provides a fine setting of the voltage within the selected range.
I just downloaded the manual, and learned it was designed such that the supplies could be safely connected in series or parallel, so increasing its versatility. I’ve never risked making such interconnections (without approval) since I saw someone do that, then had to watch their nice new power supplies emit clouds of smoke.
Farnell L30-T Front
Although it didn’t occur to me at the time, I should have taken some pics of the inside while I was checking the cable before replacing the grommet. At least I thought enough to point the camera at the rear, showing the fuses, current limiters, and voltage selector.
It’s practically like new, from the days when everything was built out of components you could actually see and hold in your hands, unlike the tiny, dust-like SMD components common today, barely visible on a massive multilayer PCB amidst a few surface mount semiconductor packages containing the equivalent of a room or two full of electronics from the past, and needing a cooling fan capable of producing a local Force 10 gale to prevent them from bursting into flames.
Today, even the fuses are tiny, dust sized SMD components that need a magnifier to see, let alone find, and switched mode power supplies mean voltage selectors are a thing of past, as most will operate from below 100 V AC to above 250 V AC with no need for adjustment.
It may have been years ago, but I still remember opening the first piece of equipment to have such a supply in our lab, and wondering if the label showing that wide operating voltage range was genuine. With a price tag of around £10,000 back then, I certainly wasn’t going to be the one turning that box of goodies on without first confirming with the manufacturer.
Farnell L30-T Rear
Things went wrong just after I renewed the grommet, and set the supply onto a small side table before I cleared a space at the rear of my bench.
I didn’t realise I’d placed it so near the edge of a small table, on top of a piece of rubber backed protective carpet tile.
About half an hour later I heard a thud – the carpet tile had slowly deformed under the weight of the supply, and the supply slid off the table/carpet onto the floor.
At least the drop was onto carpet, and only about 2 feet (sorry, 60 cm for tender, easily offended, non-bilingual millennials), but when I turned the thing on to check it – one of the supplies was completely dead (insert long stream of obscene language and curses here, just to get the full effect of my reaction).
Oh well, here we go again… off with the covers.
The good news appeared to be no evidence of damage, and both meter movements seemed to be OK.
I was initially aggrieved to see that the neon indicator on the front panel did not come on when the unit was switched on, and that one of the supplies was dead. That feeling got worse when I noticed the red lens on the indicator appeared to have been broken – it had been fine before this drop.
I later found this damage had, in fact, always been there, which was a bit of a surprise. The red lens still looks completely undamaged when seen from the front, but when seen from below (with the unit on its back), it’s obvious that it has been hit from below and pieces have been lost. When I checked the floor for pieces of the red lens where the unit had fallen, no bits of lens were to be found. This makes sense since the corner where the lens protrudes is effectively shielded by the casing, and a simple drop would not have broken just the underside of the lens. So, at least I hadn’t caused THAT particular piece of damage.
I checked the fuses, both looked identical, and both (apparently) tested OK for continuity across their respective fuse holders – yet one supply was completely dead while the other was fine.
I was puzzled by this, and while I had a few random stabs at checking other components, there didn’t seem to be anything wrong with the dead one.
Despite both fuses passing their first test, there was definitely no mains voltage reaching that dead supply.
The combined failure of the neon indicator, (not illuminating when the mains was ON), and ONE working supply – there was something odd going on.
Again, I checked for continuity across the fuse holders, and eyeballed both fuses, visually confirming both were intact.
That brought my fault-finding mistakes up to a grand total of TWO!
First was checking for continuity across the fuse holders while the fuses were still inserted. To be fair, this should be OK, but depends on the spec of the meter being used, and the circuit being checked (more later),
Second was accepting that the continuity result AND the eyeball check of the fuse wire within the glass fuse was adequate.
1.25 in 2 A Glass fuse
When I took the additional step of removing and checking the fuse from the dead supply), it FAILED. It proved to be open circuit, despite the wire inside looking perfectly normal (and identical to its partner from the working supply).
In fact, the fall had caused the fine (500 mA) fuse wire to fracture under one of the metal caps on the glass fuse, so, despite appearing complete, had a hidden break. The pic is actually a 2 A fuse – a 500 mA sample barely had visible wire inside.
I confirmed this break under a magnifier, where it was possible to see the wire vibrate at one end (the failed end) when the glass envelope was trapped.
Replacing the broken fuse brought both the second supply AND the dead neon back to life.
And that neon was the second mystery.
According to the circuit diagram, the mains indicator neon is just wired directly across the mains, after a double pole switch connecting live and neutral to the two fuses, and then the two transformers.
There’s no reason, from that diagram at least, for the neon not to remain lit if a fuse blows (or breaks) on either one or both of the independent supplies.
I can only assume (yes, I KNOW I shouldn’t do that) the published diagram (or at least the version I could download) differs from the actual ‘as built’ supply I have.
I’m just happy the fuse was the only casualty.
(For the trolls – Yes, I DID cut the grommet fit it, rather than dismantle the mains cable. GET OVER IT!)
Some extra techie goodness
Having spent a couple of days with the power supply mocking me from the back of the bench, I was compelled to drop the covers once again, to have a quick look inside to see if there was any obvious reason for the odd fuse readings, and behaviour of the mains indicator neon.
Irritatingly, I found that the good old Fluke 77 reads ‘continuity’ for quite high values of resistance, in this case 122 Ω. I may be wrong, but there appears to be no value given in the spec sheet I checked, yet I know modern multimeters generally stop signalling ‘continuity’ above 50 Ω. Still not exactly ‘continuity’ (for a wire or connection), but well below 122 Ω.
That higher figure was the resistance I found when I checked for an expected open-circuit between the output sides of the two fuse holders (before I’d thought about it properly).
Clearly something common to both, and NOT as shown on the relevant part of the circuit diagram, shown below.
Farnell L30-T Mains
At the time, and slightly misled by the diagram only showing ONE supply, it took a moment or two’s thought (and a quick scribble of a twin supply circuit) to make it obvious that the two (apparently) independent fuse holders were actually connected through the primary windings of the two mains transformers, which shared a common connection. While the transformer’s live terminals are independent, and fed from each fuse holder, their 0 V terminals share a common connection direct to the neutral wiring within the supply. Interestingly, despite being identical, one transformer was marked GEC, and the other ERG (who I’ve never heard of). They differed electrically as one primary measured 72 Ω, while the other was 50 Ω – making a total of 122 Ω (as measured above).
And the behaviour of the neon?
I followed the wiring from the protected/fused side of each fuse holder, and found it led to a piece of tag strip bolted to the chassis This had three resistors soldered to it, together with some more wires leading off… somewhere.
These appeared to lead to the neon, which unfortunately terminated in shrouded wires, hiding their colour.
After poking around this, it looked as if one led to the mains switch (neutral), while the other headed towards that terminal strip (which we already knew was connected to live via the fuse holders).
It looks as if they split the neon dropper resistor into three, each measuring about 45 kΩ. One in series with the neon, and the other two connected to that one, and fed from each fuse holder.
If either fuse fails, the current from one alone is insufficient to hold the neon ON after it strikes – a behaviour I did witness when I first tested the unit after the fall. The neon glow cannot sustain with the current supplied by just ONE of those resistors connected to the supply from the fuse holders.
A neat and simple means of indicating one of the mains fuses has failed if one of the supplies is not working.
Secret Scotland 