EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: TERRA Operative on November 08, 2018, 04:41:05 pm
-
I'm looking to replace some old circular ceramic capacitors on a bit of test equipment.
The caps are on the BNC jacks, between the isolated shield and chassis as seen in the attached images (You can see one is chipped, hence the replacement)
Now, I can find piles of 50V circular ceramic caps in Akihabara, but not much else...
My question is, can I just use something like this? https://www.digikey.com/product-detail/en/tdk-corporation/FA26C0G2A223JNU06/445-180656-1-ND/ (https://www.digikey.com/product-detail/en/tdk-corporation/FA26C0G2A223JNU06/445-180656-1-ND/)
If not, what would the best substitute capacitor type be? I don't think this is overly critical (Hence the use of circular ceramics) but I just don't want to be way off the mark...
-
Why are you replacing them?
Ceramic capacitors tend to be stable within their original specifications and are very reliable.
edit: Sorry, missed your comment about one being chipped. I'm sure it still works fine and you could just touch a dab of epoxy over the chipped area to seal and protect it, though obviously you could just replace them I suppose.
For cost and size reduction most ceramic capacitors are now a multilayer design rather than a single disc, except for ones that are rated at higher voltages where the physical ceramic layer needs to be thicker so it is more practical to make the traditional disc style. You can still find lower voltage versions, though, just less commonly.
For example, Vishay D Series (http://www.vishay.com/docs/28549/dseries.pdf) or S Series (http://www.vishay.com/docs/28515/sseries.pdf) (datasheet), eg. D203Z59Z5UL63L0R (https://www.digikey.com/product-detail/en/vishay-bc-components/D203Z59Z5UL63L0R/BC5117-ND/2832223) or S223M75Z5UN83J0R (https://www.digikey.com/product-detail/en/vishay-bc-components/S223M75Z5UN83J0R/1277PH-ND/2356813) at Digikey.
-
You should be able to use any ceramic cap with > 100V rating and reasonable capacitance. (22nF?) These just decouple noise; no one in their right mind would use Y5V caps for their signal properties. Disk or not doesn't matter.
-
Is the pulse load rating of those disks different then modern components?
If they are used in the traditional manner, directly after a wire run, can you actually replace them with a component that has similar or better durability in a different form?
What comes to my mind is surface area. I mean their decay-vs-overload compared to smaller parts. I would imagine they dissipate heat better and have a larger area so the field density is lower.
Also, how is the overload/transient durability of different dielectric grades (i.e. z series vs the supposedly traditionally higher end X series)?
-
Cool, looks like the ones I selected would work, although a bit overkill, the ones drussel found seem closer to the original specs/shape (with higher rated voltage).
I'm not sure pule load rating is too much of an issue for these caps? They are just used for coupling the shields of the test leads to the equipment chassis.
How would I determine what that value is for the original parts if it is a critical variable?
-
I'd say that the capacitance is .022uF=>22nF
-
- It is unlikely that the value of the capacitor changed as a result of the slight chip. It is probably within the original tolerance of the nominal value.
- It is unlikely that the exact value of the capacitor is critical in that particular part of the circuit.
- It is unlikely that the voltage rating is critical in that particular part of the circuit. If the voltage gets anywhere near even 50V you will have much bigger things to worry about than that capacitor.
Unless you intend to use this in a high humidity environment (like on a ship or submarine?) it may be better just left alone. If you are that worried about, seal up the chip with a dab of epoxy, etc.
-
As those are wax dipped, the simplest fix for them is a drop of candle wax on the capacitor, and a quick blast of hot air to melt it into the capacitor body again. Will reset the aging of the capacitor, but as it is a disc ceramic and will drift anyway, and as it has a -20% +80% tolerance, and will drift in this range anyway, resetting the aging with heat will not have any real consequence on use.
You might want to resolder the diode though, as it is going to cause problems with offset voltages, as the diodes are there to decouple hum, and the capacitor is there to reduce high frequency noise, so in use anyway will only have 0.6V across it maximum.
-
Is it intended that back-to-back diodes would go on each BNC? At most you'd see 1.2V across the capacitor(s). Just about anything would work.
-
Practically any ceramic or film capacitor will work in this application and the value does not need to be exact; a 0.015uF or 0.027uF would work just as well. As KD4CCF pointed out, a lower voltage part is acceptable because the diodes limit the voltage.
A 0.022uF 50V 10% X7R multilayer ceramic capacitor is probably about the least expensive option and better than the original dielectric.