Well, you can probe the die during operation, checking the various nodes. This method is very tricky however, requires delidding a hermetic package device (gold/tin lid seal solder is very brittle, a single edged razor blade and a 6 inch length of 1/4 dia stainless as a hammer work quite well at popping the lid)..Oh, and you cannot drink any coffee for half a day before probing. edit: Used to be a 60X binocular scope was adequate, I guess now you need a metallurgical scope and an actual probe station.
Or, you could ask the chip designer..
Common sense dictates that if an output is overloaded significantly beyond the design, there will be internal nodes that "bottom out".
jn
I'm sorry, but my memory isn't that great. We had them because Wayne
was using them for housekeeping on a preamp, so I set up one as a line
stage and it didn't sound good. I wanted to like it, and I screwed around
with it for a couple of days, and then moved on. Its THD measured OK.
Was it the "error correction block" or the phase splitter driving the NPN
output stage, or just me? I don't know.
Fair enough, thank you Nelson.
If I as an amateur can't remember it all, it's hardly surprising that you as a professional don't keep it all in the head.
Manual probing is pretty much gone these days. I fondly remember the tungsten needles and three axis manipulators. FIB is mostly what we use, far beyond the budget of DIY.
All I remember is those needles were sharp. The girls tried to keep me away from sharp objects, but alas, I never learned.
We purchased all kinds of silicon in waffle packs, and had to verify specs in die form. I'd put together a probe card, the girls would tune it to the pad geometry, I'd work out the electronics around it, they'd probe the dice. The best was when the waffle pack was carbon loaded for esd, we could probe in the waffle pack using the pack as backside connection. Otherwise, they had to load the dice onto a gold plated hybrid package lid.
jn
Does anyone have a copy of Dick Burwen's 1962 200W amp article?
A lot of power for early solid state.
EDIT - http://americanradiohistory.com/Archive-Audio/60s/Audio-1962-Nov.pdf
A lot of power for early solid state.
EDIT - http://americanradiohistory.com/Archive-Audio/60s/Audio-1962-Nov.pdf
Last edited:
FIB is pretty cool Nelson and I had a chance to see a serious DIY guy with $ who started doing his own digital chips. He refurbished a used machine.
Thanks for the files, Karl, I'll start having a poke around.
In the meantime, here is a genuine glitch in the original LT2 files - B and D differ obviously - that I'm looking at. Both DiffMaker and Audacity picked it up, and it's occurring at the peak of one of the highest amplitude swings - what's the likely culprit?
Last edited:
That's interesting I would think the monster FPGA's you can get now would cover anything you would want.
Yes, we've used these in several projects and they are very fast. Maybe it's the learning curve.
Yes, the adds are even more fun. I spent some time doing research today trying to find out where the open-loop BW stuff came from. Seems that Greiner published this idea back in 1964 and Otala took it up. Burwen of course was always "can't have enough feedback".
Since I know Dick and Bob maybe a clarification "from over the years of experience" could appear in Linear Audio.
Last edited:
Scott, thanks for the link to that magazine. I can't remember the last time I saw ads for Eico, Heathkit and Bell Labs. The ad for the Empire 880 was amazing, $47.50 was a lot of money in those days. The claim of over 30 dB separation caught my eye. I wonder if they were truly able to reach that spec?
That's what I said, I think the IDEA is wrong. BTW they published a IEEE version of their amplifier .6A of class A bias pretty extreme, and the VAS was loaded by 1.1K to a gain of only 11.
Maybe in a narrow area of the midband, with a carefully tweaked stylus.
If you like those, maybe you'll like this: Magazines 1952-1962 | Vintage Vacuum Audio
Last edited:
Calling Greiner wrong you might see me eating my shoe like Werner Herzog. No problem lots of hot stuff will help.
Doc G is still kicking but retired.... Madison Wis. Lets email him and 'talk' to him about his thoughts. He might have more to add than what he published long ago. He could clear up a lot of things we debate here.
Maybe even an interview for L.A.? There is an old African proverb --- "When a person dies, a library burns to the ground".
THx-RNMarsh
Maybe even an interview for L.A.? There is an old African proverb --- "When a person dies, a library burns to the ground".
THx-RNMarsh
Last edited:
It must be said that it was Daugherty who wrote his PhD thesis with this concept of using high open loop bandwidth. Greiner was only his advisor. Later Dr. Greiner became very conservative about audio with his wire articles, etc. I doubt that he will defend high open loop bandwidth today.
Dick Burwen was a brilliant circuit designer in the early days. Yet, he shifted to the HA909 IC by the early 70's, just like I did originally. This was the best IC available back then, but it still had sonic problems. Its slew rate of +5/-2.5V/us was marginal, especially as we used +/-24V supplies and 600 ohm loads on occasion. What I did NOT know about the RA909 (911) was that many individual samples had very high crossover distortion. Yet others were pretty good. Apparently Dick Burwen sorted these parts and put the good ones in the audio path and the others in signal processing applications.
The very first preamp that Mark Levinson built was with the assistance of Dick Burwen. It was just as beautifully made as any product that Levenson ever made, had meters, was very expensive, and YET got a so-so response from the audio reviewers like TAS. If what many of you think about audio reviewers was true, they would have jumped on this new product, and put the audio research tube stuff aside, but they didn't.
That is where I came in (1973) and gave Mark some discrete designs to try. He heard the difference and every IC was ultimately replaced by my discrete designs.
Of course, today, Mark is long gone from Mark Levinson, the company, and they WILL now use IC's and not know any difference, but it is 40 years later. The slew rates have gone up with the later devices. Hopefully the crossover distortion has been greatly reduced as well.
Are the best IC's equal better or even as good as the best discrete designs remains to be proven. I haven't seen it proved yet.
Dick Burwen was a brilliant circuit designer in the early days. Yet, he shifted to the HA909 IC by the early 70's, just like I did originally. This was the best IC available back then, but it still had sonic problems. Its slew rate of +5/-2.5V/us was marginal, especially as we used +/-24V supplies and 600 ohm loads on occasion. What I did NOT know about the RA909 (911) was that many individual samples had very high crossover distortion. Yet others were pretty good. Apparently Dick Burwen sorted these parts and put the good ones in the audio path and the others in signal processing applications.
The very first preamp that Mark Levinson built was with the assistance of Dick Burwen. It was just as beautifully made as any product that Levenson ever made, had meters, was very expensive, and YET got a so-so response from the audio reviewers like TAS. If what many of you think about audio reviewers was true, they would have jumped on this new product, and put the audio research tube stuff aside, but they didn't.
That is where I came in (1973) and gave Mark some discrete designs to try. He heard the difference and every IC was ultimately replaced by my discrete designs.
Of course, today, Mark is long gone from Mark Levinson, the company, and they WILL now use IC's and not know any difference, but it is 40 years later. The slew rates have gone up with the later devices. Hopefully the crossover distortion has been greatly reduced as well.
Are the best IC's equal better or even as good as the best discrete designs remains to be proven. I haven't seen it proved yet.
That is what he is currently doing for his latest project to be built by TSMC. The simulation models are rather large and quite analog like. I would imagine AD has some nifty tools like that.
- Status
- Not open for further replies.