|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
97.95.43.235
In Reply to: RE: How about a10KHz sq wave w/1uF across the load R? (nt) posted by Steve O on August 30, 2016 at 18:19:26
10khz sq wave with 1uf across the load.
" Not too shabby I guess.
"For every complex problem there is an answer that is clear, simple, and wrong" H. L. Mencken
Follow Ups:
...I doubt many modern tube amps would perform as well. In fact thst is so good I suspect an internal LP filter is in circuit somewhere...or rise time of signal is long.
When I rebuild the 55a amps they will be much better I could guess.
"For every complex problem there is an answer that is clear, simple, and wrong" H. L. Mencken
Ha, ha, no, don't do it. LOL
--------------------------
Buy Chinese. Bury freedom.
across it so I'm sure of the fundamental of the cap in this case?
"For every complex problem there is an answer that is clear, simple, and wrong" H. L. Mencken
The cap adds a reactive component to the load that more closely approximates a real world speaker, especially an electrostat. Most feedback amps will exhibit overshoot and ringing to some degree or another when presented with a reactive and esp a capacitive load. The overshoot and ringing are indicative of overall amp stability. The less the overshoot and more damped the ringing, the more likely the amp is stable under typical operating conditions. Some amps will spontaneously oscillate w/o applied signal at some ultrasonic frequency when presented with a reactive load.
I always check for stability when working with a vintage amp to be sure it won't blow something up if high cap cable or estat type speaker is anticipated.
Looking at that 10KHz waveform again, it looks suspiciously bandwidth limited. Next time you check sq wave perf, I suggest you put the scope in dual trace mode and show both input to amp as well as output.
The bandwidth limited doesn't surprise me because I did a sweep from 20hz to 30khz and you could see a rapid drop off after about 16khz.I think the amp needs to be rebuilt because the power supply is not up to the task when you get 11% distortion at 52 watts.This amp has phenomenal iron and at the very least we should be relatively flat to 30khz..Just looking at how good the 22khz sq wave was on the Mc225 after the McShane type rebuild,that dictates the importance of the power supply because it definitely increased the bandwidth in that amp by quite a bit.This is not to say it was lousy when new,but when it came to me it was barely making its rated power and distortion was 40% over spec at full power. It does have a surprisingly symmetrical clip even when you see the notch distortion.BTW,I can change that pattern with the control on the amp which affects the FB loop as I recall.I'll look at the schematic again.
"For every complex problem there is an answer that is clear, simple, and wrong" H. L. Mencken
Edits: 08/30/16
Some amps are specified pretty close to their inherent limits. Meeting those specs now may require all electrolytics and coupling caps to be renewed, and for tubes to be new and meet the same specs they did back in the '50s or '60s. Also, it's worth considering that some amps were rated for output power using a methodology other than continuous RMS. Short term power measurements can result in higher numbers because the power supply doesn't sag during the tests.
--------------------------
Buy Chinese. Bury freedom.
I always do sweeps and square wave checks at close to full power as opposed to 1 watt as most manufacturers do. Had I done these tests at 1 watt,the numbers would have been much better.I usually do the test at 1 watt as well but I do it after a power supply rebuild.
"For every complex problem there is an answer that is clear, simple, and wrong" H. L. Mencken
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: