Forums

Old Anelex power supply: AC on the DC

Started by Terry Pinnell August 25, 2018
This huge, ancient power supply made by Anelex has been serving from
under my shed workshop bench for about 25 years:
https://www.dropbox.com/s/g11s67bb5jk7bc8/HD-Power-Supply.jpg?dl=0

It supplies a few projects around the house and garden and I generally
just ignore it. But a couple of days ago I found that there is 115V AC
between the case (connected to the 230V AC mains input) and all the DC
outputs. Could have more than a bit to do with the shock I described in
my earlier post 'DC cable puzzle'!

What are the more obvious causes to pursue please?

Terry, East Grinstead, UK
On Sat, 25 Aug 2018 08:21:12 +0100, Terry Pinnell
<me@somewhere.invalid> wrote:

>This huge, ancient power supply made by Anelex has been serving from >under my shed workshop bench for about 25 years: >https://www.dropbox.com/s/g11s67bb5jk7bc8/HD-Power-Supply.jpg?dl=0 > >It supplies a few projects around the house and garden and I generally >just ignore it. But a couple of days ago I found that there is 115V AC >between the case (connected to the 230V AC mains input) and all the DC >outputs. Could have more than a bit to do with the shock I described in >my earlier post 'DC cable puzzle'! > >What are the more obvious causes to pursue please? > >Terry, East Grinstead, UK
Water ingress, or corrosion in the power transformer... The transformer is the only place for that to happen, BUT completely disconnect the output from the load to make sure it is the supply and not the load picking up stray leakage from somewhere else. The primary is probably arranged so you can feed from 120/60 or 240/50. So a wiring error like grounding the two primary windings at the midpoint would give you exactly 1/2 your mains supply V and those symptoms you mention. If a load resistor between the 115 you are seeing and case drops the voltage to zero that may indicate a water or corrosion malfunction (suspect those same terminals) - but start with a light bulb not a tiny 1/4 watt resistor (lest if vaporize between your fingers. A hard (wired) connection will not drop the voltage much and should cause a 240 V incandescent bulb to glow dimly. Stay safe. Observe good safety practices, this nuisance has the potential to turn lethal. BTW that supply looks like something we used in the 60's in the Navy, and there's no reason it shouldn't last longer than your shed if it's in a dry protected location.
default <default@defaulter.neo> wrote:

>On Sat, 25 Aug 2018 08:21:12 +0100, Terry Pinnell ><me@somewhere.invalid> wrote: > >>This huge, ancient power supply made by Anelex has been serving from >>under my shed workshop bench for about 25 years: >>https://www.dropbox.com/s/g11s67bb5jk7bc8/HD-Power-Supply.jpg?dl=0 >> >>It supplies a few projects around the house and garden and I generally >>just ignore it. But a couple of days ago I found that there is 115V AC >>between the case (connected to the 230V AC mains input) and all the DC >>outputs. Could have more than a bit to do with the shock I described in >>my earlier post 'DC cable puzzle'! >> >>What are the more obvious causes to pursue please? >> >>Terry, East Grinstead, UK > >Water ingress, or corrosion in the power transformer... The >transformer is the only place for that to happen, BUT completely >disconnect the output from the load to make sure it is the supply and >not the load picking up stray leakage from somewhere else. > >The primary is probably arranged so you can feed from 120/60 or >240/50. So a wiring error like grounding the two primary windings at >the midpoint would give you exactly 1/2 your mains supply V and those >symptoms you mention. If a load resistor between the 115 you are >seeing and case drops the voltage to zero that may indicate a water >or corrosion malfunction (suspect those same terminals) - but start >with a light bulb not a tiny 1/4 watt resistor (lest if vaporize >between your fingers. > >A hard (wired) connection will not drop the voltage much and should >cause a 240 V incandescent bulb to glow dimly. > >Stay safe. Observe good safety practices, this nuisance has the >potential to turn lethal. BTW that supply looks like something we >used in the 60's in the Navy, and there's no reason it shouldn't last >longer than your shed if it's in a dry protected location.
Thanks, very helpful. Continuing testing today. Terry, East Grinstead, UK
On Sun, 26 Aug 2018 08:33:46 +0100, Terry Pinnell wrote:

> Thanks, very helpful. Continuing testing today.
I had the same thing happen with an old Advance RF sig gen I owned about 40 years ago. It was fine for years then suddenly started outputting 115V on the sig-out socket that was supposed to be 1mV-1V! Blew a lot of transistors before I discovered the fault, 'cos I have very high skin resistance and couldn't feel it. One particularly dull day I noticed a tiny spark jumping from the probe tip to the DUT and on further investigation discovered the high V output. The Earth connection to the metal case should have prevented this, but the previous owner (a ham radio operator) had for some bizarre reason inserted an insulating bush between the Earth tag and the chassis! This may or may not help you but it just reminded me how stupid some (supposedly intelligent) people can be. -- This message may be freely reproduced without limit or charge only via the Usenet protocol. Reproduction in whole or part through other protocols, whether for profit or not, is conditional upon a charge of GBP10.00 per reproduction. Publication in this manner via non-Usenet protocols constitutes acceptance of this condition.
On Sat, 25 Aug 2018 08:21:12 +0100, Terry Pinnell wrote:

> This huge, ancient power supply made by Anelex has been serving from > under my shed workshop bench for about 25 years: > https://www.dropbox.com/s/g11s67bb5jk7bc8/HD-Power-Supply.jpg?dl=0
Strange design having all that heat sinking inside the case and so close to those caps. :-/ -- This message may be freely reproduced without limit or charge only via the Usenet protocol. Reproduction in whole or part through other protocols, whether for profit or not, is conditional upon a charge of GBP10.00 per reproduction. Publication in this manner via non-Usenet protocols constitutes acceptance of this condition.
On Sun, 26 Aug 2018 08:52:36 -0000 (UTC), Cursitor Doom
<curd@notformail.com> wrote:

>On Sun, 26 Aug 2018 08:33:46 +0100, Terry Pinnell wrote: > >> Thanks, very helpful. Continuing testing today. > >I had the same thing happen with an old Advance RF sig gen I owned about >40 years ago. It was fine for years then suddenly started outputting 115V >on the sig-out socket that was supposed to be 1mV-1V! Blew a lot of >transistors before I discovered the fault, 'cos I have very high skin >resistance and couldn't feel it. One particularly dull day I noticed a >tiny spark jumping from the probe tip to the DUT and on further >investigation discovered the high V output. The Earth connection to the >metal case should have prevented this, but the previous owner (a ham >radio operator) had for some bizarre reason inserted an insulating bush >between the Earth tag and the chassis! >This may or may not help you but it just reminded me how stupid some >(supposedly intelligent) people can be.
There are valid reasons for "floating" the sig gen from ground from a troubleshooting point of view (like ground loops) but safety and trouble shooting expediency can be at odds. One of those 3 wire into 2 adaptors makes more sense so you can float it when necessary without wiring changes. Or isolation transformers...
On Sun, 26 Aug 2018 05:28:21 -0400, default wrote:

> There are valid reasons for "floating" the sig gen from ground from a > troubleshooting point of view (like ground loops) but safety and trouble > shooting expediency can be at odds.
Ah, well maybe that's what he did, then. Just forgot to remove the spacer after he'd finished doing whatever he was doing with its internals. Easy mistake to make. -- This message may be freely reproduced without limit or charge only via the Usenet protocol. Reproduction in whole or part through other protocols, whether for profit or not, is conditional upon a charge of GBP10.00 per reproduction. Publication in this manner via non-Usenet protocols constitutes acceptance of this condition.
On Sun, 26 Aug 2018 12:36:08 -0000 (UTC), Cursitor Doom
<curd@notformail.com> wrote:

>On Sun, 26 Aug 2018 05:28:21 -0400, default wrote: > >> There are valid reasons for "floating" the sig gen from ground from a >> troubleshooting point of view (like ground loops) but safety and trouble >> shooting expediency can be at odds. > >Ah, well maybe that's what he did, then. Just forgot to remove the spacer >after he'd finished doing whatever he was doing with its internals. Easy >mistake to make.
In a chemistry lab I worked at they had these tablet dissolution or disintegration machines. Anyhow, lots of pumped electrolyte (hydrochloric acid to simulate conditions in the stomach. Big metal cased machine with the works electronics as well as plumbing in one chassis on a lab bench (with sinks and other grounds nearby). The old geezer that'd been running that test most of his career swore he got a shock from a machine... sure enough that whole lab bench was wired with the hot and neutral switched around, and the case of the machine was connected to neutral not ground - so I dig out the schematics and damn if it wasn't showing that on the schematic: neutral going to chassis and ground connected to neutral. But like most accidents, several things have to go wrong before someone is hurt, it wasn't until the benches were rewired that we discovered the instrument ground problem as well as the bench wiring problem. Another time I watched a woman lab rat (affectionate name for the chemists that worked there) get shocked transferring water from a plastic spigot to a plastic jug. I didn't believe it, but of course checked it out. Turns out the DI water was so pure it was a perfect insulator (yeah water can be an insulator, just not tap water) and we had inadvertently created what was Millikan's water drop experiment... The friction of the water in the plastic piping was imparting a static charge and the 3 gallon plastic vessel she was filling was picking up and storing a hefty static charge. I put a brass pipe fitting on the spigot and grounded the fitting and that fixed it.
On Sun, 26 Aug 2018 08:56:16 -0000 (UTC), Cursitor Doom
<curd@notformail.com> wrote:

>On Sat, 25 Aug 2018 08:21:12 +0100, Terry Pinnell wrote: > >> This huge, ancient power supply made by Anelex has been serving from >> under my shed workshop bench for about 25 years: >> https://www.dropbox.com/s/g11s67bb5jk7bc8/HD-Power-Supply.jpg?dl=0 > >Strange design having all that heat sinking inside the case and so close >to those caps. :-/
It looks like a mil-spec supply so may be over-kill on the heatsinks, and it may also be old enough to have germanium transistors which don't suffer heat as well as silicon. Note the fine tuning on the voltage outputs? That suggests it was made before they began incorporating remote sensing of voltage to eliminate voltage drop between supply and load. Or 1960-70 or so. The pass transistors may be PNP germanium devices.
On 26/08/18 15:17, default wrote:
> On Sun, 26 Aug 2018 12:36:08 -0000 (UTC), Cursitor Doom > <curd@notformail.com> wrote: > >> On Sun, 26 Aug 2018 05:28:21 -0400, default wrote: >> >>> There are valid reasons for "floating" the sig gen from ground from a >>> troubleshooting point of view (like ground loops) but safety and trouble >>> shooting expediency can be at odds. >> >> Ah, well maybe that's what he did, then. Just forgot to remove the spacer >> after he'd finished doing whatever he was doing with its internals. Easy >> mistake to make. > > In a chemistry lab I worked at they had these tablet dissolution or > disintegration machines. Anyhow, lots of pumped electrolyte > (hydrochloric acid to simulate conditions in the stomach. Big metal > cased machine with the works electronics as well as plumbing in one > chassis on a lab bench (with sinks and other grounds nearby). > > The old geezer that'd been running that test most of his career swore > he got a shock from a machine... sure enough that whole lab bench was > wired with the hot and neutral switched around, and the case of the > machine was connected to neutral not ground - so I dig out the > schematics and damn if it wasn't showing that on the schematic: > neutral going to chassis and ground connected to neutral. But like > most accidents, several things have to go wrong before someone is > hurt, it wasn't until the benches were rewired that we discovered the > instrument ground problem as well as the bench wiring problem. > > Another time I watched a woman lab rat (affectionate name for the > chemists that worked there) get shocked transferring water from a > plastic spigot to a plastic jug. I didn't believe it, but of course > checked it out. Turns out the DI water was so pure it was a perfect > insulator (yeah water can be an insulator, just not tap water) and we > had inadvertently created what was Millikan's water drop experiment... > > The friction of the water in the plastic piping was imparting a static > charge and the 3 gallon plastic vessel she was filling was picking up > and storing a hefty static charge. I put a brass pipe fitting on the > spigot and grounded the fitting and that fixed it.
I once touched the metal front lid of one of our instruments in a rack, and got a minor shock. On checking, the instrument's metal case was obviously and measurably connected to the mains protective earth. The person who had built it said the shock was impossible, but still refused to touch it (coward!) :) Eventually we noticed the front lid's hinges were plastic, so it was actually floating with a minor induced charge. Quickly cured by a wire bypassing the plastic hinge.