His is a neverending story of "I discovered this 30 or 40 years ago", and you drooling amateur slobs haven't even learned what I knew back then.Gives him a chance to tell his story ........
The incessant put downs are his defense mechanism. He needs to gain some confidence. And he needs to stop living in the past. We've gotten past 741's and tantalums. And DA, the newest? sigh, how many decades ago?
Where's the martini emoticon when you need it??
jn
OK, some more recent 'history'. About 35 years ago, a guy named Richard Marsh (same guy as here) who worked at LBL at the time, (JN is not the only guy here who worked in a major lab for some time) started to connect DA in capacitors to potential audio problems. He published an LTE or two in TAS, which was then much like the level that we discuss here, at the time.
Then Richard and Walt Jung got together and wrote a seminal paper together in 'AUDIO' in 1980, that convinced me (at least) that avoiding coupling caps was the only good way to improve performance, and relatively easily done, BUT what about EQ caps?
Now, a few years later, I got some input from a couple of guys, who you know: Walt Jung and Scott Wurcer
THEY had come up with a differential subtraction test that compared an 'ideal' cap with what another cap, and you could measure the difference, if you used the right test signal. This test signal, (not invented by me) was a bandwidth limited pulse with a width about 10th the scope rep rate where the pulse would repeat, over and over. This test signal brought out all kinds of changes, up to 10% between the best and the worst caps.
Scott Wurcer kindly supplied at the time, the AD524 instrumentation op amp that he designed, and a Spice simulation showing virtually the same results as what we were measuring when DA was dominant in the test cap.
That is when they handled the 'ball' to me and I tested every kind of cap that I could get hold of. I made 100's of tests, and ultimately I submitted to Walt Jung a representative sample of my measurements, and together we published a paper on the test, first in TAS, and then in HFN. We got lots of static from that, some of which is laughable (to me) and we might get into that later.
(more later)
Then Richard and Walt Jung got together and wrote a seminal paper together in 'AUDIO' in 1980, that convinced me (at least) that avoiding coupling caps was the only good way to improve performance, and relatively easily done, BUT what about EQ caps?
Now, a few years later, I got some input from a couple of guys, who you know: Walt Jung and Scott Wurcer
THEY had come up with a differential subtraction test that compared an 'ideal' cap with what another cap, and you could measure the difference, if you used the right test signal. This test signal, (not invented by me) was a bandwidth limited pulse with a width about 10th the scope rep rate where the pulse would repeat, over and over. This test signal brought out all kinds of changes, up to 10% between the best and the worst caps.
Scott Wurcer kindly supplied at the time, the AD524 instrumentation op amp that he designed, and a Spice simulation showing virtually the same results as what we were measuring when DA was dominant in the test cap.
That is when they handled the 'ball' to me and I tested every kind of cap that I could get hold of. I made 100's of tests, and ultimately I submitted to Walt Jung a representative sample of my measurements, and together we published a paper on the test, first in TAS, and then in HFN. We got lots of static from that, some of which is laughable (to me) and we might get into that later.
(more later)
As you can see here:
http://www.analog.com/library/analogdialogue/cd/vol4n1.pdf
the uA715 was here at least since 1970. Notable exception among contemporary opamps. It needed care regarding freq. comp. and PCB design, but, you got gain + speed at the same time, that was what we needed, gain, speed and high SR.
I did not know your papers on cap distortion, in 1979 (frankly I had not checked amateur electronic magazines much). I new Otala + yours papers on TIM, but maybe I have had a little bit different view - like when IC is slewing it cannot amplify at the same time. It is completely occupied by slewing and no place for linear action. So, we needed high SR + input RC to prevent slewing in any case. In fact, same approach, expressed in different words. Frankly, I had never tried TIM method those days, but always took care to prevent slewing.
Last edited:
I greatly appreciate what you, your company, + several other companies (mostly closed nowadays), have done in the field of analog design and research.
George, would you kindly be more specific, I probably do not understand quite well your question.
SR of the uA715 depends on gain for the reason of different freq. comp. at different gains. So, it has higher slew rate at higher gain, so the full output swing at gain 100 is extended to some 800kHz, though it is only about 100kHz at gain 1. Is this the answer to your question?
Last edited:
I am not sure what you mean. I know very well properties of the uA741, but I could not use it for our purposes even in the year 1979. It was just too slow.
Frankly, I was very surprised, in 1979, why any of audio companies used it in their products. In 1979, I considered all audio products as commercial, using the cheapest components possible, and I was highly underestimating this branch of EE industry. Maybe I was wrong, but maybe not.
Huh? MC1741 was Moto's 741 same 1/2V/usec. slew rate.
http://www.allxref.com/search.htm?part=1741S
John have you ever thought there plenty of folks that knew this stuff and kept it to themselves? I saw feedforward 301's in the 60's, not that bad actually.
Last edited:
I can't even remember how I came by mine now. It was either a leftover from my days as a TV repair tech although I can't for the life of me think of anything that used a 741S or it was supplied "in error" from the days when I used to order lots of stuff to play around with as a youngster from cheapo companies such as Bi-Pak. But fast it was. I discovered that when playing around with opamps and squarewave testing them (as you do when its all new to you
) and finding it as good as a TL071 (from memory). I have an even vaguer recollection of a 741AS ?- Status
- Not open for further replies.