DIY Audio Projects Forum
http://diyaudioprojects.com/Forum/

Help! heavy oscillation occurred within my amplifier circuit
http://diyaudioprojects.com/Forum/viewtopic.php?f=10&t=6193
Page 3 of 3

Author:  ILoveHiFi [ 08 May 2018, 00:42 ]
Post subject:  Re: Help! heavy oscillation occurred within my amplifier cir

I don't live in bay area sorry, crossover was measured using standard oscllie scope, allot of that crossover like distortion can be very easily seen by eye.

I listen to my amplifiers instead of beliving on THD, I think this is more accurate and better.

I think its better to pm each other otherwise we are getting way off topic

Author:  Yungman [ 08 May 2018, 01:44 ]
Post subject:  Re: Help! heavy oscillation occurred within my amplifier cir

Too bad you are not in the Bay Area.

Ha ha, I think OP is long gone already.

Back to OP, if I were to do opamp power amp, I'll look into this one

https://www.digikey.com/product-detail/en/texas-instruments/OPA551PA/OPA551PA-ND/295612

This is a +/-30V opamp that is reasonable priced, this will support about 40W @8ohm or 80W @4ohm with good heatsink.

Or a little harder to use SMD +/-70V part

https://www.digikey.com/product-detail/en/linear-technology-analog-devices/LTC6090CS8E-PBF/LTC6090CS8E-PBF-ND/3838541

This one can do 100W @8ohm.

These are NOT audio quality opamps, BUT NEITHER is MC1458.

Author:  Yungman [ 08 May 2018, 01:48 ]
Post subject:  Re: Help! heavy oscillation occurred within my amplifier cir

KochiyaYamato wrote:
I listen to my amplifiers instead of beliving on THD, I think this is more accurate and better.


I found there is a lot of corelation between THD and sound quality with my speaker. My speaker is one of those hard to drive. I found very low THD, very low output impedance amp, using very low series impedance speaker cables make a big difference in sound quality.

Normal THD are published using 8ohm resistor load. This is not accurate and representative at all.

Author:  ILoveHiFi [ 08 May 2018, 17:08 ]
Post subject:  Re: Help! heavy oscillation occurred within my amplifier cir

Same here with higher gain op amps which means much more feedback.
I had approx reduced distortion by 200times because I had tried increasing openloop gain by approx 200x .
Sound quality had increased and music sounds much more warmer like valves,I don't have aTHD meter though

The open loop gain is arround 92db at low frequencies, possibly higher. Hand calculated, not using sim because finding it inaccurate。
It has three stages of voltage gain.

Author:  Yungman [ 08 May 2018, 19:33 ]
Post subject:  Re: Help! heavy oscillation occurred within my amplifier cir

KochiyaYamato wrote:
Same here with higher gain op amps which means much more feedback.
I had approx reduced distortion by 200times because I had tried increasing openloop gain by approx 200x .
Sound quality had increased and music sounds much more warmer like valves,I don't have aTHD meter though

The open loop gain is arround 92db at low frequencies, possibly higher. Hand calculated, not using sim because finding it inaccurate。
It has three stages of voltage gain.


Me and you are the very few that agree THD is very important. My experience is the lower the THD( has to be with very low output impedance and very good speaker cable) is very important to get good sound stage, 3D and separation.

My first amp is a text book design from Bob Cordell and Doug Self, with active load on the input differential pair, darlington VAS and triple EF OPS. All to increase the openloop gain. I got the THD published in the book. Most of the commercial amp don't have all the bells and whistles, they all leave out some of most. My bet is stability concern, they don't want to deal with it.

It was not easy to tame the amp, took me 2 weeks to tame it and passed all the stability test. That's where I said LTSpice is not accurate and cannot predict stability at all. I worked on a lot of RF discrete circuits, closed loop feedback control system designs, I had to resort to all that to tame the amp. But after it stabilized, it was beautiful.

I found one more very important factor on top of high openloop gain, low output impedance. Low THD measured with load resistor doesn't mean a whole lot, need very low output impedance to control the speaker. This is because speaker is non linear, amps with ultra low THD but not very low output impedance will measure THD a lot higher if the load is speaker or other non linear loads. This is where many pairs of output transistors shine, there's no way out of it. Openloop gain drops at higher frequency, output impedance of the OPS is the only saving thing. That's the reason I use 9 output pairs run at 110mA each. The damping factor without closed loop is 280, above 2000 to 4000 at audio frequency with closed loop feedback.

I have the Nakamichi and an Acurus 3X200, their THD are higher, they just don't have the transparency, separation at all. Not even close.

I have plan to design a tube push pull amp. I want to listen to something that has higher THD to make my conclusion. I bought a single ended Tube amp kit and built it. It sounded terrible with my speaker. My speaker is known to be hard to drive, that amp just didn't do it.

I am still curious about Nelson Pass. His amps are extreme high end. I can only say his Stasis amps are not good as I own one. I am still curious in his Aleph. I have plan to buy a kit on ebay and build a quick and dirty one to try. From LTSpice, THD is awful, stability is awful!!! But those amps are in $5K to $10K range. I have to conclude I am missing something.

I am using a QuantAsylum analyzer, it's a computer spectrum analyzer. It's around $400, works really well. It can test below 0.001%THD.

Author:  ILoveHiFi [ 08 May 2018, 21:56 ]
Post subject:  Re: Help! heavy oscillation occurred within my amplifier cir

I used lt spice and it was poor. Maybe more expensive simulators will hopefully work well
But I'd rather do trial and error to get optimum settings for compensation capacitors. Fruther more I do all op amp internal designs on paper and using a calculator. This way works best for me.

I found that by reducing the distortion theres much more detail and sounds to be heard, as well as sounding more warm. Every thd that exist takes away all those lovely fine details.

Some guys might even think that reducing the closed gain by resistor ratio to set feedback would decrease distortion, I would say the distortion reduction is too small to make a diffrence if done this way.

Author:  BaseMell [ 19 Feb 2019, 13:04 ]
Post subject:  Re: Help! heavy oscillation occurred within my amplifier cir

Hi..as per my knowledge RF2 should be made to be 820 ohms or lower. This means c1 arround 100 to 820uF. This allows internal capacitance to charge up faster.
680ohms and 100uf or so should be fine. I had my amp oscillate on me before when I made RF2 680ohm and c1 from 470uF to 1000uF.
RF2 can be directly grounded however there is dc gain and dc stablity of amp is reduced. Grounding via cap means zero dc gain

Author:  ILoveHiFi [ 21 Feb 2019, 01:48 ]
Post subject:  Re: Help! heavy oscillation occurred within my amplifier cir

RF2 with 680r works best when c1 is 330uF.
470uF is ok but is on the verge of being too large.

If a amp has input capacitor and fair dc stability the capacitor can be shorted scraficing negletible dc stabilty

Page 3 of 3 All times are UTC - 6 hours [ DST ]
Powered by phpBB® Forum Software © phpBB Group
http://www.phpbb.com/