Roberto, that amp is only from circa 2003, so the e-caps are effectively quite modern. If those caps have aged to the point of replacement then that indicates the amp has had a hard life and been in a hot environment. Also not helped by the 56k balancing resistors if they sit on the cap terminals. But maybe the caps were still fine and hadn't degraded for ESR, capacitance or leakage current.
That amp also appears to use the same 50+50u cap for all stages. You indicate they all had similar ESR at the two test frequencies. Were you able to test for capacitance as well?
A 50uF 500V cap will likely have a falling (capacitive) impedance through to circa 1kHz, and maybe even to 10kHz depending on how low the ESR is. Also this appears to be a can cap, where leads are taken to can rim and terminals, and so their higher frequency impedance would be altered at some stage, just because of measurement setup and intrinsic lead lengths.
Note that e-cap ESR various a lot with operating temperature - so just keeping a cap cooler, or hotter, will be just as significant as adding a fixed resistor to modify the ESR.
A valve amp like the 2203, using guitar speakers, is not likely to provide a test setup for noticing the difference in e-cap performance if the capacitance of test caps is made the same.
That amp also appears to use the same 50+50u cap for all stages. You indicate they all had similar ESR at the two test frequencies. Were you able to test for capacitance as well?
A 50uF 500V cap will likely have a falling (capacitive) impedance through to circa 1kHz, and maybe even to 10kHz depending on how low the ESR is. Also this appears to be a can cap, where leads are taken to can rim and terminals, and so their higher frequency impedance would be altered at some stage, just because of measurement setup and intrinsic lead lengths.
Note that e-cap ESR various a lot with operating temperature - so just keeping a cap cooler, or hotter, will be just as significant as adding a fixed resistor to modify the ESR.
A valve amp like the 2203, using guitar speakers, is not likely to provide a test setup for noticing the difference in e-cap performance if the capacitance of test caps is made the same.
Thanks trobbins, just one comment: the schematic shown in the previous page is the reissue of the original one built since 1975. That specific one should be dated late seventies. Those caps were around 45 years old. IIRC (I don’t have the data with me) it has 100u on first node, then 3H 120mA 110Ohm to 50u for screens, then 20k to 50u for PI, etc...
Capacitance was within +-10% of the nominal value, maybe a lucky set of caps considering usual tolerances.
Capacitance was within +-10% of the nominal value, maybe a lucky set of caps considering usual tolerances.
You can add a very small series resistor, like 0.01 Ohm to hear the difference. No harm done and cheap.
A few cm of cable can do the trick as well.
A few cm of cable can do the trick as well.
Ok I can see that schematic for JMP. Even caps from that era, or earlier can show quite low ESR (at 10kHz where I measure).
The most recent can cap I tested was a 3x 24uF 500V from 1969 made guitar amp - the caps filtered everything past the screen supply - they all measured the same 28uF capacitance, and leakage current at 450V and ESR of 1.54 ohm at 23C, although the first cap was 5% higher. A circa 1990's 24uF 500V had 1.2 ohm ESR. That indicates that 50uF 500V from 1970's may have a nominal ESR around 1 ohm, so any experimentation with series resistance may just need to bring any 'low ESR' cap up to circa 1 ohm.
The most recent can cap I tested was a 3x 24uF 500V from 1969 made guitar amp - the caps filtered everything past the screen supply - they all measured the same 28uF capacitance, and leakage current at 450V and ESR of 1.54 ohm at 23C, although the first cap was 5% higher. A circa 1990's 24uF 500V had 1.2 ohm ESR. That indicates that 50uF 500V from 1970's may have a nominal ESR around 1 ohm, so any experimentation with series resistance may just need to bring any 'low ESR' cap up to circa 1 ohm.
Thank you very much trobbins and Lampie519,
I will test 220 mOhm, 470 mOhm and 1 Ohm in series first, if anything can be heard.
Probably not, as you were explaining me in previous posts.
trobbins, thanks for all the time you dedicated to such a newbie question and thread.
I will test 220 mOhm, 470 mOhm and 1 Ohm in series first, if anything can be heard.
Probably not, as you were explaining me in previous posts.
trobbins, thanks for all the time you dedicated to such a newbie question and thread.