Numatic George pump lead open circuit

[Bob Eager]

Went to use George to clean a carpet this morning. The pump didn't work.

Checked continuity and found that the lead from the switch to the pump appears to be open circuit on the 'live' (brown) side.

This lead is two separate stranded wires with Lucar style connectors and an outer loose fabric sleeving. No sign of damage.

Pulling back the outer reveals that the brown wire is sleeved in black for part of its length and has little hard cylinders in it. Suppressors?

I don't want to just replace the wire without knowing what I'm leaving out.

Anyone?

[Bob Eager]

Aha. Seems it's a diode. Anyone have any more info?

[Andy Burns]

Probably cheaper than finding a suitable replacement from Maplin, if you can wait a day or two ...

[Bob Eager]

Yes, that's how I found out. I have ordered one from somewhere.

Meanwhile, it's a 1N4007. Which isn't a 'thermal' diode in that sense, so not quite sure how this works.

I probably have one somewhere. Off to check.

[Bob Minchin]

What do you mean by a thermal diode? The forward voltage of any diode has a temperature coefficient -0.002V per degree C for silicon)but that is the only thermal connection to a diode I can think of.

I presume the pump in question is run from half wave AC to control the speed or power dissipation for some reason.

Some cheap electric drills have a switchable diode to provide two speeds/power levels rather than doing it properly with a gearbox.

[David Lang]

The Gotek pump in those works on the same principle as an electric door bell. Would that be anything to do with it?

[Graham.]

Are you sure it's a diode? If it's the size of a typical diode and bullet shaped with a full metal jacket (as it were) then it's a thermal fuse.

[Bob Eager]

I'm just repeating what the spares list says on one website! Which I think is wrong. It's just a diode, probably for the reasons you give.

[Bob Eager]

Since it has a plastic casing and has '1N4007' written on it...

[Bob Eager]

The pump clearly works on AC as there are no contacts. But it may be half wave AC as described.

Anyway, I'll replace the diode and see what happens.

[Archibald Tarquin Blenkinsopp]

If the pump manufacturers recomendations are followed then one "bump" is a fuse and the other a diode.

The fuse is easy to check and the diode normally fails short, i.e it will read low ohms both ways round on an Ohmeter.

Numatics parts list makes interesting reading. A Numatic George built from manufacturers spares must approach the cost of a small mansion in Belgravia.

AB.

[Archibald Tarquin Blenkinsopp]

Sorry, forgot the link!!

AB.

[Bob Eager]

Turns out two of the 'bumps' are crimps and the third is the diode.

I agree!

[Bob Eager]

_Technical_data.pdf

Thanks...interesting.

[Andrew Gabriel]

Sounds like one I've had turn up at a repair event. IIRC, one of the crimps was a bad connection, and the diode had overheated as a result. I think the motor was a DC (permanent magnet) motor. I don't think a 50Hz AC motor that small can be made which is powerful enough - not even enough space for having universal motor field windings.

[Bob Eager]

The pump is a strange device. Basically a winding surrounding a spring loaded metal slug in a cylinder, with a non return valve and a channel connecting the two ends of the cylinder (can't see the channel but it must be there).

No chance to work on it, probably until the weekend. Need to find some crimps.

[David Lang]

The work on the same principle as an old fashioned door bell. Made by Gotek if that's any help.

[Bob Eager]

Yes, someone posted a link. Except doorbells don't need diodes, and only AC doorbells!