Hi Mart-
I never encountered the behavior you mentioned in 1172. My amps oscillations were triggered by increasing the input signal beyond a certain level. Your oscillation seems to be sustained and occurs with no input. My understanding is that increasing the driver base resistors is a strategy for eliminating parasitics and that's what it did for me. I don't know if it's likely to cure your oscillation but it's worth trying. If you have changed the feedback ratio to reduce gain this might be contributing to the problem. In that case you might want to increase Cdom. Finally, the changes I listed in my earlier post give a gain of about 26 and the amp is stable with a Vq of 55mv and the original Cdom.
Regards
Steve
Post 1294 , appears to me poor square wave response , no ..? I thought you had made changes to cdom ...? Are my assumptions Incorrect..?
Regards
Regards
Well I guess it's a matter of what you consider poor. The mod. amp appears to have better square wave response than the original for whatever that's worth. I measured the slew rate of the mod version using Self's method and got approximately 25v/usec. I didn't make that measurement on the original. That seems fast enough given that I'm only looking for about 120 watts out.
What are you looking for in terms of square wave response?
Steve
What are you looking for in terms of square wave response?
Steve
I was not able to carry over the pics , but would you care to go over screen shots from left to right , square waves are not square and show a lot of slewing ,
(Rounded edges ) the clipped sine waves look unsymmetrical , my interpretation could be wrong , is it possible for you to express what you are measuring per screen shot and expected results ...
Regards ,
The screen shots are labelled with the S. wave frequency and whether they were made on the original or modified MK III. I posted these mostly because you requested them and I run the test routinely as a quick check on amp freq. response. No amp is going to have perfect square waves at high frequencies unless it has infinite bandwidth. The rounding off you mention is a consequence of the amps limited BW and not in this case evidence of a defect. Both amps have a BW that is more than adequate for reproducing audio signals. They both also have input high and low pass filters that limit their BW.
Yes both amps show asymmetrical clipping. There is also evidence of "sticking" on the clipped negative peaks. I don't know how significant either effect is if the amp is not routinely driven into clipping (would like to hear others opinions on this). In any case I don't know how to correct either "defect."
Hi Bonfis
Wanting to take this opportunity to say thanks for posting and sharing in an attempt in tackling this , aside , I'm not able to tell frequency of square wave posted, I had originally asked if possible for you to post squares at 20, 1k, and 20k , the sticking affects sonics as it affects recovery , playing with bias and comparing to see a difference may tell us something.
Only one of the posted square waves looks acceptable , there is enuff experience here to make this amp better or correct for the lack of a better term , best we compare stock as per Carlos to modified mk3 and take it from there ...
Regards ,
Wanting to take this opportunity to say thanks for posting and sharing in an attempt in tackling this , aside , I'm not able to tell frequency of square wave posted, I had originally asked if possible for you to post squares at 20, 1k, and 20k , the sticking affects sonics as it affects recovery , playing with bias and comparing to see a difference may tell us something.
Only one of the posted square waves looks acceptable , there is enuff experience here to make this amp better or correct for the lack of a better term , best we compare stock as per Carlos to modified mk3 and take it from there ...
Regards ,
Last edited:
OK left to right as follows: Original MK III 10 Khz, 30 Khz, 40 Khz, Clipping
Modified MK III 10 Khz, 20 Khz, 40 Khz, Clipping.
BTW the Original was measured with Carlos' recommended bias while the Modified was measured with about 57mv bias. Both have the same problem with sticking so I don't think bias changes will help that.
IMHO there is nothing about the square wave response that suggests a problem. If you disagree please tell me why.
Hello Bonfis,
Now that its a lot clearer to me as to what the graphs posted represent , i had thought the 40K was 20 K and 10 K as 1 K , so you are correct , about the square wave response , could you post Sq waves @ 20 hz and 1 k ..
IMO i would like to see less filter for a bit more bandwidth than is there currently , makes a big difference with the spatial perception of the amp, interesting to hear what others think regarding this , the bandwidth is better with the modified version ...
Is it possible for you to post graphs @ 1 k and 20 hz if possible, also are you running matched transistors in the outputs ...?
Now that its a lot clearer to me as to what the graphs posted represent , i had thought the 40K was 20 K and 10 K as 1 K , so you are correct , about the square wave response , could you post Sq waves @ 20 hz and 1 k ..
IMO i would like to see less filter for a bit more bandwidth than is there currently , makes a big difference with the spatial perception of the amp, interesting to hear what others think regarding this , the bandwidth is better with the modified version ...
Is it possible for you to post graphs @ 1 k and 20 hz if possible, also are you running matched transistors in the outputs ...?
Last edited:
The 1 Khz trace is perfectly square. The 20 Hz trace shows some tilt as expected due to the input high pass filter. Yes the transistors are matched.
The input filter does not determine the bandwidth of an amplifier it only limits the frequency range of the signal reaching the LTP.
Hi friends 🙂 I look for Dx Blame MKIII and i think that is great amplifier ... but i can't find the complete documentation about this amplifier ..
Can you Carlos give me the full documentation for Dx Blame MKIII because i want to buld it but to be in some pdf file ? 🙂
Regards , Daniel 🙂
Can you Carlos give me the full documentation for Dx Blame MKIII because i want to buld it but to be in some pdf file ? 🙂
Regards , Daniel 🙂
Last edited:
- Status
- Not open for further replies.