Unknown part, photo

If memory serves (and it may well not) that's an old 'wet-slug' tantalum capacitor. If so, corrosive stuff inside.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

On Tue, 27 Sep 2011 02:35:56 -0400, Spehro Pefhany put finger to keyboard and composed:

It looks like a tantalum to me as well. The date code is suggesting the parts were made during week 42 of 1994, but I recall seeing these during the early 1980s. In fact they used to fail catastrophically and burn holes in the PCB.

- Franc Zabkar

--
Please remove one 'i' from my address when replying by email.

YEs, that looks like a Cap.

25V, 50uf +/- 10%

I would replace that if it isn't reading, it most likely is connected to the unijunction oscillator.

btw, those relays are prone to coil burn out. We use a lot of them in the plant and I've never had one cook a unijunction how ever, we've had bad caps and bad coils in them.

P.S. some of ours used an IC, you must have an old type or one that has a short timing spanned on it.

If you had a scope with isolation or a battery operated scope this would be much simpler to diagnose. You would read between B1 and B2 of the unijunction of pulsing activity.

Jamie

Tantalum capacitor. Thanks gang!

I imagine the red end is the equivalent of the neg marking on an electrolytic (ie connect to more negative voltage (or gnd)).???

I have now read that reversing the leads can lead to capacitor failure, I think I messed it up in my testing trying to determine what it was. I don't think it holds a charge anymore but I was seeing it at one time.

Sad. Will replace with a modern tantalum.

Alp

Red usually indicates positive.

You'll likely find 50 uF difficult to get . Use 47 uF instead. If this is for your Potter Brumfield relay circuit, it is likely that you don't need a tant - you could probably use an aluminum electrolytic instead. Generally, tantalums cost more, so at least you could experiment with cheaper caps and replace with tants after you've sorted out the polarity issue.

Ed

As for the polarity issue, the negative normally goes to ground, so determine the ground. However if it's a coupling cap, then your guess is as good as mine.

BTW: What piece of equipment is this thing in?

Well if red is positive that might explain why my part substitution is not working. I thought I would put the neg end of an (47 uf) electrolytic cap where the red end of the tantalium was. I'll verify on the working "delay/relay timer" which end the red end was at and try again on the broken one. Might yet get it working.

Alp

A timer relay. see previous post in this thread for the exact type if you are interested.

Alp

Your posts on this, for some reason, made me want to dig a lot deeper, looking at some of the posts you made elsewhere about this issue. Some of the information I found was not readily apparent to those who have responded here, so I'll add it into this post. If I mis-state any of the findings, please correct the errors. Here's what I've found:

You want to repair a Potter & Brumfield CUF-41-30120 time delay relay. The exact model of the relay is of critical importance. The datasheet (referenced below) does not show a CUF-41-30120. It does show a CUF-41-70120, and it also shows a CUH-41-30120. The reason the exact model is critical is that the model number indicates the relay coil voltage, which could be 24 volts DC or 24 volts AC, or 120 volts AC. Knowing the correct coil voltage is important for us to figure out what circuit is inside the relay case. It may be that CUF-41-30120 *is* the correct model and is not shown on the datasheet, but it is worth double checking.

You have already replaced the relay, which restored your machine to proper operation, thus proving that the relay is indeed defective. You want to repair the defective relay for use as a spare, rather than building an add-on (external to the relay) circuit. Purchasing a new relay is too costly: ~$200

You suspect two internal components: MSU112-2 which someone told you is a ujt and the 50uF tantalum capacitors. You say you might have installed the caps backwards. Both of these issues need to be double checked. A search for a datasheet for MSU112-2 or MSU 112-2 or MSU112 or MSU 112 does not find one. It may be that the transistor is not a ujt - without a schematic or a datasheet we can't be sure.

The datasheet for the relay can be found by clicking on "Compact Time Delay Relay" near the bottom of the page at:

Per that datasheet, your relay is a "Delay on operate ? Delay period begins when input voltage is applied. At the end of the delay period, the relay will operate and will not release until input voltage is removed." (I am assuming the datasheet applies because the model numbers shown there are *so* close to the model # you posted.)

An adjustable delay on operate timing circuit can be made with 5 components: a transistor, a potentiometer, a resistor, a capacitor and a zener. (If you don't need an adjustable delay, you don't need the pot - instead, you use a fixed resistor.) You already have the pot as part of your existing broken relay. (I'm assuming yours has an adjustable delay.) The question is: do you have room inside the relay case for those parts if you remove the existing parts? The above also assumes that your circuit uses 24 volts DC to energize the relay.

Ok, please check the model # you have against the datasheet info. Measure the coil voltage that the machine provides to energize the relay to see if it is AC or DC, and the voltage level. Examine your broken relay to see if the parts mentioned above could fit in the available space. And please post the results. :-)

Regarding your original circuit, the best thing, if possible, is to repair what you have. We know that the existing components fit in the available space, and there is no need to create a new circuit if the defective component(s) can be identified and replaced. Examine that transistor very closely to see if maybe the part is labeled something other than MSU 112-2. If it is, please post the correct number. You mention a diode - are there any numbers visible on it? Many diodes have part numbers starting with

1N, like 1N4148, IN4001, 1N4746, 1N914 and so forth. See if you can determine the part number for your diode. Also, in one of your posts you mentioned that a few parts were connected inside the relay case. So far, as I recall, you mentioned 2 caps, a diode and a transistor. Are there any other parts that you didn't mention?

Finally a general comment. The price of time delay relays is really annoying. Take a 10 dollar relay, and a buck or two worth of parts, and presto! you've got a $200 relay. :-( From the manufacturer's point of view I can see it: limited volume, tooling costs, production time, etc. But from the consumer point of view ... grumble. I think that may be why I'm so interested in this "saga". :-)

Ed