Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Capacitor Replacement

Status
Not open for further replies.

Pud

Mechanical
Mar 26, 2003
2,707

Please bear with me - this is not my field....

Last year we had an electronically controlled proportional valve fail (on an injection moulding machine) - replacement cost 1700GBP. The mechanical part of the valve is fine.

Yesterday the engineer from the company gave me a service fix sheet which involved changing a failed electrolytic capacitor (0.2 GBP I guess!).

The question is, is there a "better" type of capacitor one can use? (e.g. more reliable!!) The current one is 47mfd, 50V, aluminium can type, both wires out of one end.



Cheers

Harry
 
Replies continue below

Recommended for you

Is sounds like the repair bill of my brother's giant screen TV set. Hundreds of dollars for a fuse, and a careful inspection...

A better grade of capacitor may not clear the problem, if it is caused by a design limitation. If this was the only faulty part, one thing you can do is to note exactly the brand and part number of the cap. Buy a spare one, and have a local techie replace it if your valve fails again the same way, before calling the repair guy.
 
To improve a 47uF 50V Electrolytic radial, you could
A) Replace it with a 105C rated part. Most are rated for 85C.
B) Install one rated for a higher voltage (63 or 100 volts). The higher voltage part may have a greater lead-spacing between the two leads.

However, the real issues are:
1) Why did it fail in the first place?
2) Does the fix sheet really convey the correct and full information of what went wrong?

Electrolytics primarily fail due to heat and overvoltage/transients. Heat can come from the ambient environment, or from large ripple currents internally heating the capacitor. Transients come from voltage surges, lighting, etc.


There was a rash of failing electrolytic capacitors made in Taiwan and China about 5 years ago due to several companies incorrectly copying a proprietary electrolytic formula of another company. They would fail after about a year, and this affected a lot of PC mother boards. But, this involved only a small part of the electrolytic market.
 


felixc: The 1700GBP was just for the valve - changed that myself - one plug and 4 bolts! 5 minutes!

Comcokid: I see the guilty cap is 150C (well, it's got 150 printed on it)

Your comment re Asian copies might have it - the machine was 4 years old last year when it failed and the fix sheets do not mention other types to go in. (It's also a different colour to some of the other ones on the board..) Also, we have another machine with the same valve on it which is about 18 months older than the failure and no problems - curiouser and curiouser said Alice......

Will just change for now with replacement from reputable source and try it (might of course be something different!)

btw, valve made by Bosch

Cheers

Harry

 
If the cap is an electrolytic it won't have a temperature rating of 150 [°]C. Could be a film type I guess. Any chance of a photo? Comcokid's suggestions are right on the money for solutions.


----------------------------------
image.php
Sometimes I only open my mouth to swap feet...
 

Scotty: apologies - serious typo - is 105 on it.

Cheers

Harry
 
That reminds me of a municiple water works that asked me to design a new board for a valve. They had a failure every couple of months, boards $1,100. Manufacture told them they needed to buy bigger valves. The board that was failing only was used to power a couple of LEDs. From various pieces of the two watt resistors from about a dozen boards I was able to determine it was a 10 ohm resistor. My guess was that it was supposed to be a 10K. Even the right value resistor would cook the board over time. Big meetings and blank faces saying no one else is having a problem. The manufacturer finally caved in and gave them new boards. On those I replaced the the resistor with a capacitor.

Moral is...Don't expect a pipe manufacturer to know electronics or an electronics sub contractor to care after the design is out the door.
 
Sounds like you already have a high temperature electroytic. Can you work out what it is doing in the circuit? Something to consider is ripple current rating, especially if it is in some sort of filtering or smoothing duty. Little caps don't like high ripple current, especially if the ambient temperature is already high.


----------------------------------
image.php
Sometimes I only open my mouth to swap feet...
 
Funny. No sooner that I posted my response, the lab tech came to me and said two of the lab computers were becomming flakey. We looked at the mother boards, and there were 47uF 50V caps swelling. Still experiencing the bad electrolytic issue - even after several years!

By the way - the normal aging mode of electrolytics in constant use is to slowly lose capacitance over time and temperature as the electrolytic slowly dries out.
 

Hi all,

Many thanks for the help.

I did bring the camera to work and took a photo - then realised I had left the USB cable at home - Doh!

Will get it on tonight for you to have a look at - it's mounted next to what looks like a transformer - possibly a pule transformer as it appears to have a ferrite core (two E shaped bits?) I really am guessing here......

No swelling or anything though...


Cheers

Harry
 
Hi folks,

Here it is:
The suspect is the one on the right, nearest the bottom of the picture of the group of 5 ...


If anyone in the UK can recommend a "fixer" or can volunteer, would appreciate a call on Prudhoe 836938

Cheers

Harry
 
Pud,

If it is urgent, Dowding & Mills have an excellent repair facility based out of Leeds who will collect & deliver your item back to you in a working state. Costs a little more but for those "Right Now!" jobs they can be useful. Also pretty good on PLCs and drives. Not sure what they are working over Christmas.

If it was to find its way down to Teesside Power Station in the New Year I'm willing to have a look at it for you and get the cap swapped. Testing it might be a bit trickier - no circuit diagram and so on - depends what the demise of the capacitor caused as collateral damage. There looks to be plenty of board space so putting in a cap with a better spec should be feasible. The item next to it looks like a high frequency transformer, so ripple current might be the problem. What brand are the capacitors? Sometimes the giveaway about quality is that no one outside of China has ever heard of them!

Is the splodge of blackish crud on the bottom right corner near the grey edge strip significant or is it some innocuous mark?

----------------------------------
image.php
Sometimes I only open my mouth to swap feet...
 
I like how they circle the wagons around the transformer. These little transformers are nothing but big resistors. Board layout guys always seem so obscessed with packing things in as tight as they can. Those rated hours of life for a cap, at even a moderate temperature, come rather soon when a device is operated 24 hours a day.

A bass player friend just asked me to work on his Ampeg amp. This is a recent design with tubes and FETs. They took two electrolytics and tightly sandwitched them between two hot running 10 watt resistors. Another is right up against the base of a tube. Plenty of space to put them somewhere else.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor