I just re-did all the R's and C's in the preamp section of an aged Dynaco SCA-80. It was interesting to measure the resistors to see how they had changed.
The design has mostly half-watt 5% carbon comp resistors, with a few 10% resistors. I measured them all, and in comparison to their original values, they all had gone high. It ranged from 4% to 14% high. On average, the resistors were 8.24% higher than their nominal values. I measured 34 resistors.
The design has mostly half-watt 5% carbon comp resistors, with a few 10% resistors. I measured them all, and in comparison to their original values, they all had gone high. It ranged from 4% to 14% high. On average, the resistors were 8.24% higher than their nominal values. I measured 34 resistors.
I made essentially the same observation, over many years, BUT, there are a few exceptions too: some remain rock stable, and some go low.
This is valid for resistors working in normal conditions (or NOS ones).
When overstressed, they can go either direction: high or low, sometimes very high (~open) or very low (~short).
The nice thing about that "normal" type of evolution is that the circuit continues operating almost normally, because all resistors go high in almost the same ratio, meaning the bias etc are not too upset. Unfortunately, there is often the oddball throwing a spanner in the works, and that's when you have to replace them all, because not a single one has remained within its tolerance limits...
They are also quite immune to high-voltage pulses (ESD) as well, which is not necessarily a given for film-type resistors (especially laser-trimmed precision ones, apparently).
I've heard and read the opposite. Military CC R's may be significantly different than non-military. CC's will blow open much easier than CF or MF. Sometimes you might want that. CC's are also the noisiest (hiss). Metal film is the way to go, especially for input stages, unless you like hiss. I know people who think CC sounds better... I think the additional hiss may be causing a psychological effect there.
Funny, I found the CC resistors in my 1961 build dynaco ST70 and PAS2 to be within tolerance, in 2011. Dynaco must have changed suppliers by the time ST80 was shipped. Chicago built Hammond organs 64-68 are rock stable too unless the resistor case is burned. Every H100 I've worked on has a burned 10k 1W resistor in the amp reading 11.5 to 13.5 k. I put in 2 W RN65's. European Holland built tonewheel Hammonds have the resistors go high value.
I think the paint in resistors from production lines that produced mil-spec rcr07gxxxxx resistors took a huge improvement about 1960. B/C3 Hammonds prior to that have quite a few resistors that go high value. I suspect the mil-spec paint lines were Allen Bradley and Sprague, by the appearance of the resistors and the rcr resistors we used for JSC installation in 1977-78 that looked like that, but I don't have any real evidence.
BTW the over 100k metal film resistors I put in the PAS2 and ST70 did make the hiss less in both.
David A. of NC who repairs amps said that Fisher & Scott hifis used the stable carbon comps in the early 60's, too. He says the guitar amp companies usually used the ones that would go high value after some decades.
I think the paint in resistors from production lines that produced mil-spec rcr07gxxxxx resistors took a huge improvement about 1960. B/C3 Hammonds prior to that have quite a few resistors that go high value. I suspect the mil-spec paint lines were Allen Bradley and Sprague, by the appearance of the resistors and the rcr resistors we used for JSC installation in 1977-78 that looked like that, but I don't have any real evidence.
BTW the over 100k metal film resistors I put in the PAS2 and ST70 did make the hiss less in both.
David A. of NC who repairs amps said that Fisher & Scott hifis used the stable carbon comps in the early 60's, too. He says the guitar amp companies usually used the ones that would go high value after some decades.
Last edited:
Many years ago, I worked as a chemical technician, occasionally developing and making the resins for CC formulations using the Stackpole process. That would include production trials and observation on-site. There were many variations on how the resistors could be made, the raw materials and qualities used and how they subsequently performed in the field. In later years after the resistor manufacturers had closed their plants and film resistors became standard, I joined the multi-national company we had supplied and then had the opportunity to review their old process control manuals. It was interesting to say the least.
The binder in the carbon/mineral powder formulations was a special type or blend of phenolic group resins (aka Bakelite), selected for good electrical properties, low moisture absorption and stability. Ours were based largely on Cresol but for optimum stabilty, the final product really needed a protracted "post cure" which automated production lines and schedules commonly didn't permit. I'd suggest that high rel. and mil. spec products did and still do have some level of extended cure processing to keep drift down below 15%, though I would be surprised if epoxy resins were not now in use, for their superior stability.
The bulk of production in Oz was for domestic consumer goods and the stability of the 1/4 to 2W resistors was surprisingly good until they had been running 5-10 years in a tube TV set or a compact radio where there was a bit of heat. I recall my father regularly testing imported and competitors' CC resistors with drifts of >> +100%.
The mil-spec CC resistors still sold by my employer in the 90's when I departed, were all large, high rel. types imported from Allen-Bradley and sold for an awful lot more than a few cents. As only their sales agents, we didn't need to test them prior to delivery or maintain production records and I don't recall the performance specs TBH, other than that they were reliably within 5% of their original measured values and routinely replaced on a maintenance schedule anyway.
Have a look at the measurent conditions recommended here: RCR07 PDF Datasheet - Allen-Bradley / Rockwell Automation - Datasheets360.com
The binder in the carbon/mineral powder formulations was a special type or blend of phenolic group resins (aka Bakelite), selected for good electrical properties, low moisture absorption and stability. Ours were based largely on Cresol but for optimum stabilty, the final product really needed a protracted "post cure" which automated production lines and schedules commonly didn't permit. I'd suggest that high rel. and mil. spec products did and still do have some level of extended cure processing to keep drift down below 15%, though I would be surprised if epoxy resins were not now in use, for their superior stability.
The bulk of production in Oz was for domestic consumer goods and the stability of the 1/4 to 2W resistors was surprisingly good until they had been running 5-10 years in a tube TV set or a compact radio where there was a bit of heat. I recall my father regularly testing imported and competitors' CC resistors with drifts of >> +100%.
The mil-spec CC resistors still sold by my employer in the 90's when I departed, were all large, high rel. types imported from Allen-Bradley and sold for an awful lot more than a few cents. As only their sales agents, we didn't need to test them prior to delivery or maintain production records and I don't recall the performance specs TBH, other than that they were reliably within 5% of their original measured values and routinely replaced on a maintenance schedule anyway.
Have a look at the measurent conditions recommended here: RCR07 PDF Datasheet - Allen-Bradley / Rockwell Automation - Datasheets360.com
Last edited:
Carbon film also go up in value. I just replaced all CF resistors with MF in one of my Marshall 2203 guitar amps that has seen years of constant abuse and the sound sort of collapsed recently so it was time for a in-depth check what was wrong because most operatings point were off despite of new and 100% in spec tubes. I was shocked to find pretty much all resistors had increased significantly. The 100k load resistor at the cathode follower was worst at 180k(!!), followed by the 100k PI plate loads which were at 130k and 150k.... no wonder it sounded so bad. Now it's pefectly fine again.
Last edited:
I've been wondering about RNR series resistors, which are currently 'naked' (unpainted) metal film resistors sealed in a glass tube back-filled with an inert gas. One can easily see the spiral construction of the resistive element. Carbon film resistors were once used this way too. Typically gold-plated leads. These were used in extreme environments (moisture, high or extremely low pressures, and general oxidative nastiness), but were otherwise good quality metal film resistors. I've no idea if this was done with carbon composition resistors, but I sort of doubt it.
Examples of RNR resistors can be easily found on Ebay.
Interesting that bulk foil resistors can get measurably non-linear at very low frequencies; I would expect the 'naked' parts would be worse, and oil-filled (but very expensive) foil resistors would be better.
Tantalum nitride resistors are supposed to be very moisture insensitive.
Examples of RNR resistors can be easily found on Ebay.
Interesting that bulk foil resistors can get measurably non-linear at very low frequencies; I would expect the 'naked' parts would be worse, and oil-filled (but very expensive) foil resistors would be better.
Tantalum nitride resistors are supposed to be very moisture insensitive.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.