... in closed loop amplifiers.
On my point of view, it is not a matter of closed loop bandwidth but a matter of how fast feedback correction is applied. I found always better to limit, anyway, the input signals speed as low as possible (= no phase turn at 20kHz = >200kHz). Kill the RF as well.
Well, my position about slewrate is very simple, faster it is, better it is. As it is an exponential curve, hard to tell a limit
It always seemed to me that CFA sounded better than the same topologies with VFA.
Last experience was modifying a power amp from VFA to CFA, and all the guys witch had made the same mod reported the same nice improvement in their listening experience.
As the original amp had yet a > 1Mhz bandwidth and 200V/µS, it was not a question of bandwidth, but only the slew-rate change to 1000V/µs. Or something else ? It is related with transients reproduction.
I had modified several big mixing desk , using CFA, and always the change was amazing.
Major improvement at the mixing bus side (inverting mode).
Thanks to the French office of Analog Devices for their friendly help and support during all my professional life. Nice guys with personal relationship can make the life easier and the music better.
Last edited:
Esperado: Walt Jung, Matti Otala, and even I have done significant testing of worst case audio signals and HOW they effect lower slew rate electronics. We published our papers on this, in the 1970's, where are yours?
Are you telling us that 'subjectively', you find higher slew rate better?
Why can't I say then, that 'subjectively' I find that open loop operation better?
Is your 'subjectively' better than mine? If so, how?
Are you telling us that 'subjectively', you find higher slew rate better?
Why can't I say then, that 'subjectively' I find that open loop operation better?
Is your 'subjectively' better than mine? If so, how?
Trial court mania protection ?
Well, one failure part over millions ? The next replacement one will probably work decades.
Um, that's the final exam..but with blindfolds as well, on a bus travelling at high speed over mountain trails with the seal team.
Lead free solders..too hot. And whiskers, tin...
jn
During the time you were promoting yourself, drinking glasses and congratulating yourselves in the AES private club parties with your little familly, i was WORKING ! Thousand miles away.
Like Thousands of engineers in the world you never heard about.
And sorry, your papers were not the ones we studied in French schools, but Ohm law, Nyquist etc...
Where is the Curl's Law ? Never read any 'paper' under your signature..
Can-we compare the performance of an amp you designed in 70 and a cheap one of ours with the first epytaxial high speed power transistors (CFA, THD=0.05%) ? It will be funny !
My previous response was not aggressive at all. Your is, and, once again, a try to use authority and dropping names instead of technical arguments.
It is impossible to exchange with you in a friendly atmosphere.
Last edited:
Haven't noticed that much difference for consumer gear. BTW 90% of my failed CD/DVD's have been mechanical (usuallly the drawers freeze).
I will also remind you that it is the same die in SMT as the leaded one (in most/many cases) and as someone pointed of the heat pad with direct board connection can do amazing things. We have a special purpose part right now that is doing 200V p-p at 1A peak in a tiny SMT package (very low duty cycle).
BTW there is a new application related to cellular base stations that will have some amazing fallout for DIY audio applications, more later.
Last edited:
John, I mainly do military, aerospace, and medical designs that have both a long life time and have to have a zero failure rate because someones life depends on 9one of my layouts and the circuitry behind it working without failure. Sorry you are wrong or misinformed. SMD is used for these areas because it is more reliable, better for vibration etc. My main area that I get called in to desig n are SMPS's for high reliability situations where failure or excessive noise could lead to someones death, it focus's you thoughts when desining these sort of things. And many a PTH commercial design has failed in the past, there are many levels of design, from cheep throw away commecrcial to very very high reliability. Youknow people who work on hush hush stuff, ask them what the reliability rate has to be for a lot of equipement especially where it performs a life support function. A lot of my layouts HAVE to work for 15 years + without failure, so you cannot use one bad example of a product to categerise all SMD, sorry it does not carry any wieght. Just think every time you get in a plane, train or even drive a car there is a good chance depending on the make that I may have done a PCB that is critical to the safe operation of the viechicle, and when you have an operation some of the life support kit may also have passed through my hands, sobering isn't it
The limited warranty on a lot of commercial gear is down to tin whiskers, and the use of Pb free solder, thats why high rel designs still use proper solder with lead in it, though we do have to get our BGA's reballed and the other devices re-tinned depending on what the finish is on the device, did a lot of work regarding this and reliability over the last few years as I worked on a big project with one of the USA big contractors who are very hot on this stuff and will not tolerate any lead free designs at the moment, and I dont blame them. So the reliability is not down to the devices, but down to the solder finish.
Last edited:
I had previously said that feedback and open-loop topologies are TWO different ways to reach the same goals, and even defended your choices.
And i just pointed than WITH FEEDBACK, i prefer high slew-rate and that is not a concern with open loop...What the hell ?
Why do you read everything in such a paranoid way ?
Last edited:
Esperado, IF you actually looked into what you are selecting, you would find that you are PROBABLY going for the higher quiescent version of the same op amp. Slew is just secondary to open loop linearity that you get with higher quiescent current. Please study up, and don't throw some cheap 60's mid fi amp at me, as an excellent example of design.
I love the bottom terminated devices, makes handling the heat on a populated PCB much easier, and as Scott has pointed out, you would be suprised what power densities you can achieve, even the new FET packages are more efficient that the old TO style PTH devices, mount them directly onto a ceramic heat sink with the copper pattern printed on the heat sink so you solder the device to said heatsink, and you get some realy realy low thermal resistance, so the device works optimaly, and with further implementation you can embed heat pipes into the heatsink and achieve some very high power, high slew rate RF designs in quite compact spaces...
I also do audio though mainly either professional gear, or communication systems, the parts are out there...
I also do audio though mainly either professional gear, or communication systems, the parts are out there...
Last edited:
Please provide your 'excellent example of design amp' done in 70. We will simulate and compare.
Your MID-FI pretentious argue is boring.
Sorry, Esperado, my early schematics were burned up in the 91 firestorm. However, at the time, 1969, I designed a 2000W amp that was a complementary differential 4 quadrant bridge amp with 50V/50A output for Ampex Research. This was then converted to a floating balanced current source for motor drive operation.
The circuit you showed was typical of what was in 'Wireless World' at that time. My goodness, you have no idea what we were working on in those days.
BOOTSTRAPED driver stage! Quasi-complementary output. How quaint! '-)
The circuit you showed was typical of what was in 'Wireless World' at that time. My goodness, you have no idea what we were working on in those days.
BOOTSTRAPED driver stage! Quasi-complementary output. How quaint! '-)
Last edited:
Slew-rate ?
i'm not impressed by power, i had some very poor sounding Crown power amps, and i thought we were talking about audio and slew-rate effects
What was original at this time was the use of fast devices, for slew-rate, as i said. Everybody was using poor 2N3055 at this time.
This amp was using a stabilized supply with protection, and switched resistance arrays for volume and tone controls, like all mid-fi, isn't it ?
Last edited:
- Status
- Not open for further replies.