Selecting a diode for an AC solenoid

What you're missing is probably that the diode is not the problem. Even though the symptoms are the same as when the diode was the problem.

a) That a diode tests OK on multimeter, capable of perhaps 9VDC at most does not mean that it will test OK under actual load. Even operating well below its rating, it is still a 'wearing part' in this application. b) Do make sure that the pump is free of clogging and that the moving parts are not binding. If you can remove the pump safely, try soaking it in white vinegar for 24 hours. c) Do make sure that the switches/contacts are clean and free of corrosion.

Best of luck with it!

Peter Wieck Melrose Park, PA

Sounds like you don't think mysterious diode failure modes are in the cards. I've no clue what might lead to the erratic behavior seen so far, but it's time to bark up a different tree..

Thanks for writing.

bob prohaska

Most silicon diodes short as their failure mode, but occasionally we'll see one either open or get bail under load - regardless of how they test on the diode scale of a typical DMM. Try another, nothing to lose.

It may gave transformed into a 6V zehner diode. Which means it looks like a diode on the multimeter, and like a short in the circuit.

Was the original diode a 1N5407 or something else?

This was back in the late 1990's and I don't remember. It's possible the diode was physically damaged to the point the markings weren't readable. I do remember the pump coil potting cracked and I replaced the pump at some point, though maybe not at the same time.

What got me suspecting the diode was a strong but not perfect correlation between how long the espresso machine had been heating and the onset of low pump output. First cup in the morning was fast, subsequent (~20 min later) much slower. Next morning the cycle repeated.

Occasionally the first cup from a cold start was slow also, but that was relatively infrequent, maybe once a week.

Right now the machine is set up with three parallel 1n4007's of unknown quality plus two connected as crowbars. The crowbars do seem to make a difference in slowing down the water flow, which is undesirable. That they do make some difference is slightly encouraging, as crowbarring the coil would slow field decay. And, so far the pump behavior has remained consistent.

I probably should pull the pump apart and look inside, but I really can't imagine a mechanical explanation on something so simple.

Thanks for reading and replying!

bob prohaska

I thought that I was the only one who tears something apart, let's it age for 10 to 20 years, and only then tries to fix it.

Sigh. That's not very helpful. For all I know, the original "diode" could have been a thermal fuse that somewhat resembles a diode.

Do you have the manufacturer and model number so I can do some Googling for info? Taking a shot in the dark, I found this hint suggesting that the diode might be a zener.

Here's the service manual and schematic: Hmmm... See schematic on printed Page 19. Looks like the motor has a

1N4007 diode across the winding and an SCR for power switching. That suggests that it's running on DC, not AC.

Yep. Something is getting hot. The cracked epoxy(?) potting compound suggests an overheated pump coil. The zener for the Breville depresso machine drops the peak line voltage 24V. An ordinary diode in place of the zener drops it only 0.6V. Kinda sounds like it might be a hot pump coil. Got a gun type Infrared thermometer? Measure how hot the pump coil gets.

Overheating the pump coil might make the pump parts move around. My guess(tm) is you'll find some correlation between the number of minutes (or hours) you ran the expresso machine the previous night. The longer it ran, the greater the likely hood that something in the pump became mis-aligned or moved out of place. Just a guess(tm).

After 26 years, I would expect some corrosion and lime deposits.

Y'er welcome. Good luck.

A diode alone, can not change 60Hz to 30Hz.

I repaired industrial equipment for almost 40 years and it is very rare to have an intermittent diode, but it can happen.

Considering how cheap they are, I'd just replace it.

Is the solenoid DC perhaps?

Are you sure the diode isn't there just to provide DC?

I never heard of an AC solenoid that required 30hz

You're right, my error. The diode prevents doubling the 60 Hz of the line to 120 Hz in the plunger motion.

In any event, I think Bob Englehardt's analysis might be close to the truth; the problem is not be what I suspect.

After reading Bob E.'s remarks, I had another look at the machine and found what looks like a check valve that was open both ways. It's far from obvious how that could change from one cup of coffee to the next, but I took it apart, turned over an o-ring and it sealed in reverse. This valve is in series with at least two other check valves in the pump, so it isn't obvious how it might matter..... Over the next few days I'll learn if it makes any difference.

Just for completeness, here's a link to some information on the pump:

Item 6 is a 3 amp diode, with no hint why, how or when it's used. In my pump it's in series with the coil.

It should be noted that the GAS in the parts callout must refer to the fitting types, this is most definitely a water pump. Also, the outlet fitting used is item 10, so there's no relief valve nor flow regulator. The check valve I fixed isn't part of the diagram and I'm not sure why it's present at all.

Thanks to everyone for reading and replying!

bob prohaska

If you have a solenoid connected to 60 Hz AC, it will be activated 120 times a second (positive and negative going voltage). A diode will cut that in half. So from 120 Hz to 60 Hz. Not 60 to 30, but still by 1/2.