I worked at HK (Long Island) and JBL (LA) back in the late 1970's for about 1 1/2 years, until Matti Otala superseded me. It was different then, but just as crazy. Full of middle management.
I worked on the Harman automotive amp line '87-'88. The big brass ring was to get Ford Quality 1 Supplier certification. All semiconductors had to be stored in anti-static containers. I watched as the power amp IC's were taken out of the factory cardboard boxes and put into antistatic bags. I pointed this out to the floor manager and got the dog tilting head look.
I had quality circles, TQM, SPC and then six sigma. Given the individually tuned RF circuits in some critical locations we chortled at trying to apply something that works for camshaft deliveries to Ford. But, the one that annoys me and makes me glad I'm not working for a major telco with a red logo any more is the six sigma black belt. Six sigma is fine, even if I suspect Motorola unleashed it to both allow themselves sloppier margins on some part and to slow down the competition as they raced to adopt it. But it only works on large sample sizes and not on people, gantt charts etc.
Dilbert of course has it perfectly
Dilbert Comic Strip on 1993-08-25 | Dilbert by Scott Adams
Dilbert Comic Strip on 1993-08-26 | Dilbert by Scott Adams
Dilbert Comic Strip on 1993-08-27 | Dilbert by Scott Adams
Dilbert Comic Strip on 1993-08-28 | Dilbert by Scott Adams
Oldies are the best...
Those Dilberts are all priceless. I like the Ishikawa one the most.
I got a personal thank you note from Sidney when Harman passed a quality audit by Ford. He got it a little garbled, as it read "Design of Failure Mode Effects Analysis", but it was still cool.
I got a personal thank you note from Sidney when Harman passed a quality audit by Ford. He got it a little garbled, as it read "Design of Failure Mode Effects Analysis", but it was still cool.
Yes. They are cheap enough to use both opamps in the package per rail - just cascade them. I'll sketch something up tomorrow.
Same with noise, look long enough you get one that has none. Frankly there are are standard distributions for these things and any QA guy with an education should be aware of them.
Oh jeez this is bringing back memories - AEC100, DoE, 8 D process, three different lots of 77 devices, the seven Why's and indeed FMEA's and battles with the production guys and the customers
I use a split - I.e completely independent - secondary transformer with dedicated FWB and filter caps per winding. I then use the 317 reg, or 7815 depending on my mood after which I simply stack the outputs to get a split supply. The positive reg specs are generally better than the neg reg specs.
Kaizen events with sticky notes in colors on the wall, oh the joys. Walk through Gemba inspections once a week with all the managers asking what you did last week and what you were going to do this week, of course those were the same managers that ignored your request to look at something they needed to take care of.
The Dilbert cartoon with the guy clothed in yellow stickies and living off of left over meeting food was so spot on, brilliant.
I asked the question how much PSRR is required for home audio conditions. It is along the lines of inquiry for how low THD is required or how wide freq response is required or how high slew rate is required, dynamic range required, ETC.
Interesting that the ones who usually complain if I pick the number have no input re PSRR. When I pick the threshold number, I get all manner of objections and require 'proof'.
Saying make it as high as possible is like saying make the distortion as low as possible.
I presume there is also a PSRR number or equivalent (normalized?) that is minimum for a typical/normal home high quality audio?
THx-RNMarsh
Interesting that the ones who usually complain if I pick the number have no input re PSRR. When I pick the threshold number, I get all manner of objections and require 'proof'.
Saying make it as high as possible is like saying make the distortion as low as possible.
I presume there is also a PSRR number or equivalent (normalized?) that is minimum for a typical/normal home high quality audio?
THx-RNMarsh
Last edited:
No, it is not. Noise is, but that's only one variable to hit a noise target, and that noise target is referenced to a specific function in the audio chain. For example, if my microphone has 100 uvolt of noise, that's bad. If my power amp has 100 uvolt of noise, that's likely fine. Noise, not PSRR.
I can achieve that noise by balancing and combining different attributes in an overall circuit (PSRR is only one of them), and the possible combinations are infinite.
So, you're asking a meaningless question. It is along the lines of "how long should a string be?"
You're smart enough to know that the -130dB etc. THD stuff is pretty stupid. There is real work needed on the basic sources/transducers far before this point.
I did, got to 0.00x% without too much effort. Not that this is actually a technical criterion, it's niche marketing.
Solved problem in the real world.
Solved problem in the real world.
I'll confirm noise can supersede PSRR; obviously that statement comes with a host of factors.
Regulators are great, but the really difficult problems are their voltage references. Fortunately, we can filter the pee out of noisy ones so that for audio purposes the residual low frequency noise is inaudible.
But I like to make the gain stages as immune to the power supply as possible.
But I like to make the gain stages as immune to the power supply as possible.
Well, of course. But it is a lot fun to learn how to get to -130 and beyond. Its a challenge beyond just audio. And, in the end, doesnt cost more.
I do not make basic sources. Microphones. Nor speakers. That is a problem others will need to address.
There exists already great sources and speakers but few want to build, buy or sell them, unfortunately. Newer low noise mic preamps are a step forward.
So - besides SY and his obviousness --- can you or someone make an attempt to put a number on it... with supporting qualifiers if needed. If <-130dB THD is not required, maybe 120dB PSRR is not needed (for home audio) either.
THx-Richard
Last edited:
It's all connected. If you have a target for max THD or noise at the output, you can calculate the PSRR required to stay below that limit for all spurious stuff, assuming you know the junk on your supply line.
If you wish you can approach it in a rational way ;-)
Jan
If you wish you can approach it in a rational way ;-)
Jan
- Status
- Not open for further replies.