John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
Jump to Latest
#99,724
PMA said:
Relay contacts degrade with every interruption of arc. It is easily measurable. Much better to use power MOSFETs and make ssr protection. Relay contacts are a pain.
Click to expand...
PMA, I believe that there is no reason why, in daily use, a mechanical relay in a protection circuit should have to deal with high current or high voltage. The only two events where it has to perform (and wear) are short circuits of speakers line or rail voltage (short circuit of power devices or drivers) in output line. It is supposed to never happen, it's a parachute.
 
Last edited:
#99,727
Tournesol said:
PMA, I believe that there is no reason why, in daily use, a mechanical relay in a protection circuit should have to deal with high current or high voltage. The only two events where it has to perform (and wear) are short circuits of speakers line or rail voltage (short circuit of power devices or drivers) in output line. It is supposed to never happen, it's a parachute.
Click to expand...

Depending on the configuration, most output relays (series connected) may be required to deal with the full output of an amplifier. Some listeners have been known to use the previous volume control setting, when turning on their amplifier/s, with source connected and running. In my long career, I've replaced many output relays where contacts have become pitted, despite no catastrophic faults reported.
 
#99,728
john curl said:
You guys seem to miss the 'forest' for the 'trees'. Typical low cost, medium-high volumetric efficiency caps have multiple problems compared to the highest quality film caps like Polystyrene, Polypropylene, or Teflon.
DA is just part of the problem, but non-linear distortion, tempco, microphonics, and hysteresis are also factors, depending on the cap type.
Asymmetric signals are the norm in audio, not pure sine waves like we like to use for testing. DA shows itself in a real way with asymmetric signals, as a non-nullable error and that is why the test signal that Walt and I used is asymmetrical.
Is this differential comparison test perfect? NO. In fact, Richard Marsh declined to participate in the paper, because he thought the test was not ideal in depicting DA accurately. However, I saw that the test was relatively easy to implement (thanks to Scott's AD524 that is very good in this application) and that it invariably showed that 'something was wrong' with how most cheap caps handled this sort of test signal, compared to better caps, yet could look pretty good with just sine wave testing. It can be seen that the actual signal path is modified in a different way than the best caps due to DA, non-linear distortion, etc and this is what is important to note. Now, of course, ALL caps will change the waveform, if the RC time constant is set close to the edge of the input pulse, square wave, or whatever complex waveform is used, but since the 'good' caps measure similar to each other, no matter who made them, or even what material (Teflon, Polystyrene, or Polypropylene) then the 'bad' caps are made of material that has high DA and should be avoided. I suspect that even COG ceramic caps have fairly hi DA, so they might be avoided as well, especially in higher values.
This test set-up is not perfect, but we did compensate for any RC time constant differences and even some nominal internal resistivity differences that some caps have. The idea is to simply minimize the error signal by adjusting it out, if possible. The test waveform, determined experimentally by me, through testing hundreds of caps, brings out the worst case that I could, while still staying within the audio range, and it was deliberately bandwidth limited in order to minimize any TIM or slew rate aberrations, and even the effective difference between any residual inductance in the caps. Still the test is not perfect, and can be nit-picked to death, but then you have 'missed the forest for the trees'. '-)
Click to expand...


but, Polypropylene caps are high volume low cost capacitors.
 
#99,736

Attachments

  • qsc.jpg
    qsc.jpg
    38.8 KB · Views: 232
#99,740
billshurv said:
Jan was Chairman of the AES NL Section. Not sure when he handed the reins over but he will have had access to information us mere mortals don't.
Click to expand...

It's more like I was handed over chores nobody else would touch ;-)

As far as I know, there is no link between having a function in running the society, whether internationally or regionally/nationally, and being asked to review articles. Totally different things, because totally different expertise required. Although some people presumably can do both.

Note there is a kind of two-tier system. You can submit a presentation (full auditorium type or what they call engineering brief or poster presentation, which takes place in the hallways), and being published in the Journal. It is relative easy to get in the first tier (even I managed it once ;-), but to be published in the Journal your paper must be quite rigorous and following 'the scientific method' being backed up by verifiable data and repeatable results.

Jan
 
Status
Not open for further replies.
Top Bottom