I used to slice the tops off of 2N4416s to allow photonic reset strategies in charge preamps. The interesting thing was how the chips were mounted, which translated into a greater or lesser volume of borosilicate glass surrounding the gate lead: TI had the chip sitting on the flat top of the gate lead, and others squashed the gate lead over the glass and put the chip on top. These latter parts were noticeably higher equivalent noise charge, I believe at least due in part to that detail of packaging.
Kern and Mackenzie at Bell Labs I believe were the first to identify the poor dielectric quality of the glass as responsible for a significant amount of charge noise in JFET preamps. They obtained chips to begin with and packaged them in low-loss headers, and got low noise results comparable to a group at Lawrence Berkeley despite not employing photonic reset. Of course the Berkeley group got a lot more publicity for their coolness factor.
Only rational that if things can be screwed up, they can also be screwed down.
Interesting, Chinese brush painting is one of my hobbies. My teacher took us on a private tour of China (his wife is the daughter of one of chairman Mao's 4-star generals). The difference for the priveleged in China is rather amazing, one artist had a museum quality private collection of artifacts up to 6000yr. old. A second artist had a private compound that would shame many rich westerners. He painted huge (up to 12' X 16') bamboo and lotus paintings, would not fit in the overhead.
Over the last four years the fun of the trips to China has been waneing, as the westernization keeps progressing.
Please can we make it a condition of joining DIYaudio that people recite this until they believe it? And 'identical data' means precisely that: identical data.hhoyt said:
And while I am here, can we get people to understand that short-term jitter and long-term frequency stability of a crystal oscillator are almost completely unrelated? Optimising one could degrade the other, if you are not careful.
Our biggest problem was crazing of the glass with poor handling, discovered in fixing a problem at Varian (vacuum guages for implanters). Was the charge noise DA or trapped impurities?
When the AD585 S&H was characterized it took weeks to find out that the 2pF package parasitic in parallel with the external cap had so much DA it dominated the error budget.
It helps to have the best conneections in china to see what there really is - the greatest works. Without that, there is no access. I want to get to Burma fast as central region has an airport now. The Thai tell me it will/is change fast now. Seems money rules. Money begets power and power begets control... the ultimate high after you have all the money you will ever need. :-( Looks like we are going the same way as in China the 1 and the 99%... big gap there. Did you buy any art to suppliment yur own work?
Not sure.
After about a year, circa 1973, of believing poor sources (particularly, wrong information issuing from Princeton [horrors!]), I found the IEEE Transactions on Nuclear Science and a rich trove of charge amp material, which I adapted for the photodiode array preamps. Probably the preeminent guy was Veljko Radeka at Brookhaven, and he wrote some of the best papers on noise in charge preamps. He is still with us I believe and was honored not too long ago for his contributions.
But I don't recall his attribution of dielectric noise, specirfically, just the analysis of its effects.
Brad
I am not sure about this. The necessary condition is that the device is fixed to the heatsink, and the whole thing is most audible (resonates) at several kHz, close to 10 kHz.
It better have been a 576..a 575 won't get you the time of day with me..
waittaminute!!..OP amp curve tracer??? I've AC traced bonding pads to substrate using the 576, many operators overdid the U/S and drove through the pad.. But the 576 isn't an opamp tracer...what's an opamp tracer?
jn
Last edited:
It better have been a 576..a 575 won't get you the time of day with me..
waittaminute!!..OP amp curve tracer??? I've AC traced bonding pads to substrate using the 576, many operators overdid the U/S and drove through the pad.. But the 576 isn't an opamp tracer...what's an opamp tracer?
jn
Pretty rare item. Demian might be right that it was a plug-in. It did Aol, CMRR, etc. directly by displaying Vos vs. supply, output, or input as appropriate.
Yikes... folded?! I put mine in tube rolled up. But forget the tourist towns like Beijing. Go to XIAN.... up near Mongolia. No one goes there and you will get the best artists at the best price. I find the most unusual and best work from Monks and others in remote areas. Northern Laos is good too. Met a Laos-born Canadian there - studying to be a monk - he had just returned from his art show in hong kong.... Thep Thavonsouk. See Asian Art News magazine, if you dont already get it. -RNM
I had the op amp curve tracer plug in, designed for the 577, at our lab in Switzerland in 1974. It didn't do much for my modules, (too open loop linear) like the line amp for the JC-2, BUT it really showed the thermal problems of the uA741, etc. that it was meant to do. I have a 577 now, BUT I never could afford to buy the curve tracer plug in. Demian do you have a lead on this plug in? It would be interesting to try modern IC's.
- Status
- Not open for further replies.