Relay Insulation: Microsemi BR246-S0233 - Why Resistance Changes?

[sn9434]

I have Microsemi BR246-S0233. As is, the resistance between the X2 (GND) and signal line (e.g. A2) is OL. However, when I power it up (X1 = 25VDC), leave it for 5 minutes and power off, then straight away measure the resistance between X2 (GND) and the signal line A2, the resistance is now 5Mohm. This resistance gradually increases over 5 minutes or so. Why does this happen?

 

[Rive]

Do you keep it in hand while it is powered up?

 

[CWatters]

Measuring high value resistance can be tricky because the currents involved are very small. My guess is that some stray charge is messing with the values you are measuring.

 

[jim hardy]

Reverse your meter leads. If the reading goes negative and creeps back toward positive, it means your ohm meter is charging a capacitor somewhere.

 

[sn9434]

To Rive: No, I don't keep it in my hand. it's placed on a desk.

To CWatters: Yes I agree that it's very high resistance and I was concerned about the accuracy of my reading. The issue is, I have 9/10 of these that ranges from 300Mohm - 500Mohm even after power up, and only one that resistance drops significantly down to 5Mohm. I thought it could be the inaccuracy of the measurement but I repeated it many times and the result won't change. I feel like 5Mohm to 500Mohm difference can't be simply the measurement inaccuracy?Why is this particular one reading 5Mohm and not the others? That's me guessing so I am open for explanations otherwise and really appreciate your feedback.

To jim hardy: I tried reversing the meter leads. reading does not go negative then creeps back towards positive. It is always positive.

Do you keep it in hand while it is powered up?

 

[berkeman]

The issue is, I have 9/10 of these that ranges from 300Mohm - 500Mohm even after power up, and only one that resistance drops significantly down to 5Mohm.

Are they new, or did you buy them off of eBay?

Datasheet link:
https://www.google.com/url?sa=t&rct...33-datasheet&usg=AOvVaw0iFEwrhgi0HvS4I1bUscvt

 

[sn9434]

To berkeman: No, I didn’t buy them from eBay. Bought them from the franchises distributor. They are not brand new though, 5 are date code 1638 and another 5 is 1362. The 5Mohm one is date code 1638. They’ve never been used so I figuered that the risk of any insulation material degradation is very low

 

[sn9434]

1352 not 1362

 

[berkeman]

Do both sides of the questionable relay behave the same? Since you have 10 of them, can you sacrifice this one to figure out what is going on? I'd be tempted to break it open to see if there is a visual reason for this. If you break it open and find residue inside, it would likely be some defect that kept it from being sealed well.

 

[sn9434]

To berkeman: I'm terribly sorry but I'm not sure what you meant as "Do both sides of the questionable relay behave the same?". both sides meaning A-side and B-side?
And yes, I would like to open it up to see what's going on. The unfortunate thing is, there is only one "defective" relay and 9 others are fine, so I would like to leave destructive analysis to the last, and when the time comes I will probably send it off to an analysis company. I am still tempted to open one of the good ones though, just to get better understanding of its structure. I'm guessing kinda difficult because it's in a metal casing hermetically sealed. I'll try though.

 

[berkeman]

To berkeman: I'm terribly sorry but I'm not sure what you meant as "Do both sides of the questionable relay behave the same?". both sides meaning A-side and B-side?

Yes, it looks to be a double-pole, double-throw relay, right? Do both pole sides of the relay show this low OFF resistance?

From the datasheet:

And yes, I would like to open it up to see what's going on. The unfortunate thing is, there is only one "defective" relay and 9 others are fine, so I would like to leave destructive analysis to the last,

Opening the relay's plastic enclosure does not need to be destructive. Just cut the top off with a hacksaw and try not to let much plastic dust drop into the relay's electromechanical parts.

 

[Tom.G]

Do you completely disconnect the 'defective' relay from the circuit when making the resistance measurement? If not, you are probably measuring something else in the circuit.

 

[Rive]

When it is the defect of a never used component directly from distributor it is really hard to guess what the problem is. My best guess is that it got washed at some point during manufacturing and it was not completely sealed so some washing liquid got inside...
If you want to use or 'repair' it then you might try to 'bake' it if you have the right equipment: otherwise just send it back or hack it to pieces.