ES,
Of course the Toyota system of lean was started and was based on Demming's teaching SPC principals to the Japanese so long ago, but it has been developed and expanded much from that humble beginning. In my job in aerospace we were neck deep into Lean principals and application of that on a daily basis and were constantly using those principals to improve all processes, not just in some places but overall to the entire corporations thinking. What comes from that are cost savings and time savings and better understanding of required labor content. This carried over into project management and so many other places. One great thing was that this education was paid for and pushed by upper management, it starts at the top. When you are making flyaway parts there really is a demand that you get it right and failures are not easily accepted. My major program besides all the others was the Apache helicopter rotor blades, no room for failure with those parts and we had never had a failure of a rotor blade. I was in charge of all tooling on that program and that was thousands of tools to make each rotor blade. My guys were the master tool builders and tool and die makers, I was their direct manager along with quality control inspection and document control. Everything was documented and could be traced back to an individual worker at any point along the line. Looking at time studies and work flow was a constant area where improvements were found and costs reduced. This has nothing to do with waiting for failures to happen before a change would be made, that would have been much to late and someone could have ended up dead.
Of course the Toyota system of lean was started and was based on Demming's teaching SPC principals to the Japanese so long ago, but it has been developed and expanded much from that humble beginning. In my job in aerospace we were neck deep into Lean principals and application of that on a daily basis and were constantly using those principals to improve all processes, not just in some places but overall to the entire corporations thinking. What comes from that are cost savings and time savings and better understanding of required labor content. This carried over into project management and so many other places. One great thing was that this education was paid for and pushed by upper management, it starts at the top. When you are making flyaway parts there really is a demand that you get it right and failures are not easily accepted. My major program besides all the others was the Apache helicopter rotor blades, no room for failure with those parts and we had never had a failure of a rotor blade. I was in charge of all tooling on that program and that was thousands of tools to make each rotor blade. My guys were the master tool builders and tool and die makers, I was their direct manager along with quality control inspection and document control. Everything was documented and could be traced back to an individual worker at any point along the line. Looking at time studies and work flow was a constant area where improvements were found and costs reduced. This has nothing to do with waiting for failures to happen before a change would be made, that would have been much to late and someone could have ended up dead.
Last edited:
My experience with consumer product manufacturers is that the warranty service data is used to find mistakes, engineering, supplier, etc. But I have been at meetings where the bean counters were driving to cut costs by showing what they thought were unprofitably low end user complaints! (Haven't dealt with those folks for a very long time now!)
Improving quality often reduces costs, of course for life safety issues the bar is much higher and so higher costs should be in line.
to JC --
"Information is not knowledge, knowledge is not wisdom, wisdom is is not truth, beauty is not love, love is not music. Music is the best".
-Frank Zappa 1979
Happy Holidaz to all who partake -- Richard Marsh
"Information is not knowledge, knowledge is not wisdom, wisdom is is not truth, beauty is not love, love is not music. Music is the best".
-Frank Zappa 1979
Happy Holidaz to all who partake -- Richard Marsh
Last edited:
And to you RM. Frank did make a break in the chain between truth and beauty. Interesting.
I has been awhile since I have put forth my design philosophy and experience. Of course, Nelson, Wayne, and Charles Hansen should put forth their design philosophies as well, as they do as well as me in the audio design competition.
I would like to draw on personal experience and history at this point, rather than 'design rules'.
I was looking back at what makes a CLASSIC amplifier design, whether tube or solid state. Is it because of the most sales? No. Is it because of the highest cost? Sometimes. Is it because the designer really understood the nature of whatever he was designing? Probably.
Let me give some vacuum tube examples from the past (50's and 60's):
First, the Marantz Model Nine power amp, designed by Sid Smith. I first heard it 49 years ago at a hi fi show. It really was the best amp at the show. My opinion has been duplicated by many others who have evaluated it, and it holds its own today.
Second, the Marantz 10(B) designed by Richard Sequerra, Sid Smith, and perhaps others. Still holds its own today against all the competition IF audio quality is what you are looking for.
Three, the Ampex MR-70 vacuum tube analog recorder from about 1963, designed by Erling Skov. Perhaps the last really high quality commercial analog recorder. All others are compromised for convenience and cost, as well as because of ignorance of what is important in analog recording.
Only custom analog recorders have the inherent quality comparable to the MR-70.
I would like to draw on personal experience and history at this point, rather than 'design rules'.
I was looking back at what makes a CLASSIC amplifier design, whether tube or solid state. Is it because of the most sales? No. Is it because of the highest cost? Sometimes. Is it because the designer really understood the nature of whatever he was designing? Probably.
Let me give some vacuum tube examples from the past (50's and 60's):
First, the Marantz Model Nine power amp, designed by Sid Smith. I first heard it 49 years ago at a hi fi show. It really was the best amp at the show. My opinion has been duplicated by many others who have evaluated it, and it holds its own today.
Second, the Marantz 10(B) designed by Richard Sequerra, Sid Smith, and perhaps others. Still holds its own today against all the competition IF audio quality is what you are looking for.
Three, the Ampex MR-70 vacuum tube analog recorder from about 1963, designed by Erling Skov. Perhaps the last really high quality commercial analog recorder. All others are compromised for convenience and cost, as well as because of ignorance of what is important in analog recording.
Only custom analog recorders have the inherent quality comparable to the MR-70.
I don't consider the ATR-100 a 'classic' analog recorder, any more than I rate the Studer solid state analog recorders at the very top. That is why I had to 'throw away' the Studer Electronics in order to make better analog recorders for Mobile Fidelity and Wilson Audio.
Let me better explain what is 'compromised' in the ATR100:
The ATR100 was designed by Allistair Heaslett (sp), starting the early 70's. I was a 'revolutionary' engineering example, but it was compromised by the IC's used, as well as a number of other tradeoffs.
The ATR100 was designed by Allistair Heaslett (sp), starting the early 70's. I was a 'revolutionary' engineering example, but it was compromised by the IC's used, as well as a number of other tradeoffs.
The only 'tradeoff' i had to suffer, with the ATR100 was with the ceramic magnetic heads. No wear, unlike mu metal ones, but problems of cracks in the gaps that may suddenly appear.
The best transport of all analog tape recorders, with minimal wow and flutter, due to the absence of pinch roller capstan.
And the best sound and dynamic of all the two tracks tape recorder i had used (many of them, including MR-70).
The best transport of all analog tape recorders, with minimal wow and flutter, due to the absence of pinch roller capstan.
And the best sound and dynamic of all the two tracks tape recorder i had used (many of them, including MR-70).
Perhaps the attractiveness of the MR-70 was from the addition of the subtle distortion caused by imperfect tape movement across the heads, as discussed here: Q. Are wow and flutter key to that analogue tape sound?
That's just another way of saying that if we made it 100% accurate, we would never get anything done. Some engineers have more natural curiosity which gives them the energy for working out complex details. We all choose the level of complexity we are willing to contend with based on what it gives us back. Although an obvious lack of curiosity is usually a bad sign. Hobbyists are driven by curiosity and it's dubious twin, enthusiasm, so they can at times go the extra mile into the rabbit hole and drag back something really useful to the engineers.
Yes, I agree with the iterative method - a first level of approximation gives one the rough outline of a solution, and usually a couple more levels is enough to assemble a working solution, which will mostly do the job. But one can go further ... what I find very intriguing with audio design and optimisation, and ultimately very satisfying, is that there appears almost no limit - the more effort and fine tuning one puts into it, the more one yields real rewards: greater audible accuracy in the reproduction, and subjectively remarkable playback.
By nature I enjoy delving into the complex details, it's a snug fit for me - it worked for me in the computing field, and in bursts I can still apply it in this audio game ... it's a worthwhile ride ...
By nature I enjoy delving into the complex details, it's a snug fit for me - it worked for me in the computing field, and in bursts I can still apply it in this audio game ... it's a worthwhile ride ...
I mention that in the next(?) Linear audio.
Power diodes are generally run at high injection levels. At low, the exponent is qVf/2kT, whereas at high injection, it's qVf/kT. Also, at the higher currents, silicon resistivity starts to play.
Grove mentions it well in his book. I pointed it out as it can compromise the temperature dependent parameter when measuring junction temperatures.
jn
Frank this was written by someone with a very poor understanding of semiconductor physics or someone who chooses not to deal with the complexity of doing the engineering.
Please check on some of the literature on bandgap design to see just how well known every detail of the diode equation can be.
When I talk about a CLASSIC design, I mean a 'no compromise' effort, given the limitations of materials and understanding of the underlying issues at the time the design is created. I will put 2 analog tape recording stages here. First is the MR70 reproduce stage, Second is the ATR-100 reproduce stage and also the main recording stage. Look for yourself. Could you have done better at the time?
Attachments
i had, for my 16mm & 35mm recording machines. Replacing the large output resistance used to transform the voltage in curent by a passive equivalent circuit of the recording head in the feedback path. HUGE dynamic improvement !
Oh, by the way, i used OPAs as well, so you will think it was 'compromised' ;-)
- Status
- Not open for further replies.