Author Topic: Capacitors for test gear  (Read 1864 times)

0 Members and 1 Guest are viewing this topic.

Offline msquaredTopic starter

  • Regular Contributor
  • *
  • Posts: 60
  • Country: us
Capacitors for test gear
« on: July 25, 2016, 04:14:41 pm »
Hello,

I'm planning on replacing the old electrolytic capacitors in a bunch of my test gear over the next few months (Agilent, Keithley, Fluke) and based on a bit of research I think I've decided to use Nichicon UPW series for as many of the replacements as I can.

http://www.nichicon.co.jp/english/products/pdfs/e-pw.pdf

Does anyone see any issues with these capacitors as replacements?

Thanks,
-Mike
 

Online Jay_Diddy_B

  • Super Contributor
  • ***
  • Posts: 2733
  • Country: ca
Re: Capacitors for test gear
« Reply #1 on: July 25, 2016, 04:29:57 pm »
Hi,

You should be o.k. with a brand name like Nichicon, bought from a franchised distributor like Digikey.

Nichicon caps do leak, have a look at this message:

https://www.eevblog.com/forum/testgear/hp6035a-500v-5a-1kw-power-supply-repair/msg698072/#msg698072

But, the I believe that the design of the seal has been changed and this equipment was relatively old.

Changing all the electrolytic capacitors, is a good idea on equipment that you enjoy using and are planning on keeping for a long time.

Regards,

Jay_Diddy_B
 

Offline uncle_bob

  • Supporter
  • ****
  • Posts: 2441
  • Country: us
Re: Capacitors for test gear
« Reply #2 on: July 26, 2016, 01:53:07 am »
Hi

As the little graphic on the data sheet you reference shows, there are tradeoff's. If you want longer life, you go to the TT series. Even that is an over simplification. You need to know a bit about the circuit the part is running in before you can even guess what will improve it's reliability. Ideally you want to be running the cap at about 60% of it's rated voltage. You want to be sure it has a temperature rating at least 10C above the max temperature it will ever see in the gear. You would like it to be rated for about 2X the ripple current it will see in use.

On a more practical level ... you might want the leads to come out the same way they did on the original. You certainly want the part to have a size and shape that will fit in the space available. Having lead diameters that work with the original pc board is a plus ...

Lots of variables ...

Bob
 

Offline Smith

  • Frequent Contributor
  • **
  • Posts: 376
  • Country: 00
Re: Capacitors for test gear
« Reply #3 on: July 26, 2016, 06:46:09 am »
Whatever you do, don't buy from Ebay or other unreliable sources.

I usually use Panasonic FC and FK series, and Rubicon ZL/ZLH series and I never had any problems. I tend to use 105degree and higher voltage caps for all my gear if possible.

Take a look at the Keithley gear to see if it has any hot running regulators next to caps. Some Keithley stuf I fixed had room for heatsinks, but they where unpopulated. Placing them considerately lowered the temperature of the regulators, and thus the caps. It gives the regulators more mechanical strength too.
Trying is the first step towards failure
 

Offline msquaredTopic starter

  • Regular Contributor
  • *
  • Posts: 60
  • Country: us
Re: Capacitors for test gear
« Reply #4 on: July 26, 2016, 06:52:40 pm »
Thanks for all the feedback so far.

Definitely not going down the ebay route, only Digikey or Mouser.

My process has been to review the service manual parts list and build a list of components. Then do a physical inspection to verify ratings and values and gather physical attributes (radial or axial, height, diameter and lead spacing). Also to make sure there are no extra or missing components, I've seen in the past service manuals don't always match the device.

I've been only looking at 105c components with the highest life cycle possible while still meeting the physical requirements. I'll take a look at manufactures other than Nichicon, those pictures looked nasty. (even though they were pretty old)

I was thinking about posting a spreadsheet with the replacement capacitors per instrument as I work through my meters hoping to save other people from future work. :)
 

Offline uncle_bob

  • Supporter
  • ****
  • Posts: 2441
  • Country: us
Re: Capacitors for test gear
« Reply #5 on: July 26, 2016, 10:37:49 pm »
Thanks for all the feedback so far.

Definitely not going down the ebay route, only Digikey or Mouser.

My process has been to review the service manual parts list and build a list of components. Then do a physical inspection to verify ratings and values and gather physical attributes (radial or axial, height, diameter and lead spacing). Also to make sure there are no extra or missing components, I've seen in the past service manuals don't always match the device.

I've been only looking at 105c components with the highest life cycle possible while still meeting the physical requirements. I'll take a look at manufactures other than Nichicon, those pictures looked nasty. (even though they were pretty old)

I was thinking about posting a spreadsheet with the replacement capacitors per instrument as I work through my meters hoping to save other people from future work. :)


Hi

One of the things you will find is that the price of capacitors comes down as your quantity goes up. You also will find that capacitors have gotten smaller over the many years since the gear was designed. You may find that Some values make sense mainly because of the other stuff you have, rather than the specific piece of gear in front of you.

Just to be clear:

Let's say you have a supply running at +15V. It's got a 16V 47uF cap on it.  (Yes this does happen ...).

Your target cap voltage should be in the 24V to 48V range. Your target capacitance should be in the 47 to 100 uf range. If the gear is 40 years old, a modern 48V 100 uf cap likely fits with room to spare.  Buying 100 pieces of (say) 36V 100 uf caps may be a lot cheaper than buying the same quantity split over ten different values / voltages.

Bob
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf