How Closely Do The Values Of Capacitors From The Same Lot Match?

Where we discuss new analog design ideas for Pro Audio and modern spins on vintage ones.
User avatar
mediatechnology
Posts: 5437
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

How Closely Do The Values Of Capacitors From The Same Lot Match?

Post by mediatechnology »

How Closely Do The Values Of Electrolytic Caps From The Same Lot Match?

The intrinsic value matching of capacitors from the same lot sometimes comes up in design discussions of mic preamps and AC-coupled balanced line inputs. Capacitor value matching ahead of differential inputs affects low frequency common mode rejection.

Common mode rejection design limits should be based on the published tolerance but the actual relative matching of caps from the same lot are usually better. But how much better?


While recapping my Texar Audio Prisims I decided to measure the 4.7 uF axial lead caps as I installed them.
Since I was already measuring value I turned on the relative tolerance feature.
I used the first cap out of the bag to set the "standard."

48 of 59 caps, 81.3%, were within ±1%.
11 of 59 caps, 18.6%, were in the range of ±1 to ±2%.
Out of the 59 I measured none were greater than ±2%.

The value matching is much better than I anticipated.

The caps were Nichicon TVX1H4R7MAD 4.7 µF 50V axial lead that I purchased in 2018.

Your mileage may vary...
User avatar
terkio
Posts: 304
Joined: Thu Feb 07, 2019 3:17 am
Contact:

Re: How Closely Do The Values Of Electrolytic Caps From The Same Lot Match?

Post by terkio »

Very interesting.
What was the tolerance of these caps to begin with ?
User avatar
mediatechnology
Posts: 5437
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: How Closely Do The Values Of Electrolytic Caps From The Same Lot Match?

Post by mediatechnology »

They were ±20% tolerance.
User avatar
JR.
Posts: 3697
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: How Closely Do The Values Of Electrolytic Caps From The Same Lot Match?

Post by JR. »

Not surprising from production manufacturing. Chunka-chunk automated production lines won't wary widely, while tolerances specs are are mostly important for batch to batch variation. When designing for large scale manufacturing we ASSume the worst, and sometimes get it. (I won't bore you with bad capacitor stories).

I recall back in the day making a similar measurement of 5% carbon film (Rohm) resistors and they were likewise very close, IIRC much tighter than 2%.

JR
Cancel the "cancel culture", do not support mob hatred.
User avatar
mediatechnology
Posts: 5437
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: How Closely Do The Values Of Electrolytic Caps From The Same Lot Match?

Post by mediatechnology »

Just received a shipment from Mouser to replace the power supply filter caps.
I only needed 4 but I bought 10.

The value is 2200 µF 40V ±20% Vishay MAL202117222E3 (Axial)

7 of 10 were within 1%.
3 of 10 were within 2%.

I used a test frequency of 100 Hz for the 2200 µF.
On the mF scale I could only read to 2 decimal places so ±1 LSD is an error of 0.47%.

So far 70-80% of the parts in the same lot - for values large and small - match within 1%.
User avatar
terkio
Posts: 304
Joined: Thu Feb 07, 2019 3:17 am
Contact:

Re: How Closely Do The Values Of Electrolytic Caps From The Same Lot Match?

Post by terkio »

In the 60's I was interested in 2R R ladders, so I sorted in a lot of 1K 5% resistors, they came out well matched, better than I presumed.
I was told, resistors were sorted at the end of the production line, so the 5% I got were those that did not make it in the 2%.
So I had those from the same of the two sides between 5% and 2%.
Does this make sense ? Is it this Now days ?
User avatar
JR.
Posts: 3697
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: How Closely Do The Values Of Electrolytic Caps From The Same Lot Match?

Post by JR. »

I just got in 50 pieces of 33uF electrolytic caps, 20%. I only measured about ten of them but found a range of 33.5uF to 34.5uF, so roughly a 3% variation, nominally just above the specified value... not too shabby for (cheap) 20% parts. I think I'll use them. :lol:

JR
Cancel the "cancel culture", do not support mob hatred.
User avatar
JR.
Posts: 3697
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: How Closely Do The Values Of Electrolytic Caps From The Same Lot Match?

Post by JR. »

terkio wrote: Thu Sep 17, 2020 6:53 am In the 60's I was interested in 2R R ladders, so I sorted in a lot of 1K 5% resistors, they came out well matched, better than I presumed.
I was told, resistors were sorted at the end of the production line, so the 5% I got were those that did not make it in the 2%.
So I had those from the same of the two sides between 5% and 2%.
Does this make sense ? Is it this Now days ?
I expect they are now tighter coming off the production machinery, but it may vary between manufacturers.

JR
Cancel the "cancel culture", do not support mob hatred.
User avatar
AnalogJoe
Posts: 95
Joined: Mon Dec 24, 2018 3:11 pm

Re: How Closely Do The Values Of Electrolytic Caps From The Same Lot Match?

Post by AnalogJoe »

What about the tempco? heat them with a hot air gun and report back.
Bonum certamen certavi, cursum consumavi, fidem servavi.
User avatar
JR.
Posts: 3697
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: How Closely Do The Values Of Electrolytic Caps From The Same Lot Match?

Post by JR. »

AnalogJoe wrote: Fri Sep 18, 2020 4:35 am What about the tempco? heat them with a hot air gun and report back.
Speaking of temperature, that reminds me of a production problem I encountered with potentiometers. The resistive element is generally a screened on carbon ink. The tolerance of the bulk resistance is around 20% but not widely known part of the production process involves curing the ink with elevated temperature. Cooking the substrates longer can pull in out of tolerance components. At Peavey we encountered one vendor (CTS cough) who's process was so far off, that they over cooked the substrates to pull in final resistive element resistance and in the process made the parts so brittle that they would break while being inserted into PCBs. I tried to work with them, they just needed to tweak the green resistance of their ink.

Almost humorously CTS decided to no-bid, dropping us as customer, because they said we were "too picky" :lol: .

JR
Cancel the "cancel culture", do not support mob hatred.
Post Reply