The latest questions can be easily answered by both Rick and Infinite. I am in to much pain to try. Sorry.
aparatusonitus said:
1G plastic module/thick film power amps for audio
example see http://www.bmm-electronics.com/Product.asp?Product_ID=185
solution
> AD844 is an great opamp (I'm using it as well)...but it has one big flaw...it's out of production, so availability, even for diy purpose, in future is questionable.
??
At least no mention on the AD website :
http://www.analog.com/en/other/militaryaerospace/ad844/products/product.html
Patrick
??At least no mention on the AD website :
http://www.analog.com/en/other/militaryaerospace/ad844/products/product.html
Patrick
Hi Demian,
How dare you! 😉
Those thick film Sanyo things gave the service depots more business than anything else since the beginning of the transistor audio era. They were perfect! Just replace the black thing with all the legs, ECG has one that fits all.
Of course, countless "techs" still managed to destroy the pads. Then there were the guys who soldered the pins before cranking the package down. 😱 It really was wild and woolly back then.
Sanyo then really caused havoc by discontinuing popular parts on a regular interval. Just try and tell a customer that his output stage is discontinued. That was always fun, but so was using a mark IV part where a mark II used to be. They would generally oscillate, that's if the number of pins were the same on each version.
I am so glad those days are almost over. This solution is now called " a chip amp". 😀
-Chris
How dare you! 😉
Those thick film Sanyo things gave the service depots more business than anything else since the beginning of the transistor audio era. They were perfect! Just replace the black thing with all the legs, ECG has one that fits all.
Of course, countless "techs" still managed to destroy the pads. Then there were the guys who soldered the pins before cranking the package down. 😱 It really was wild and woolly back then.Sanyo then really caused havoc by discontinuing popular parts on a regular interval. Just try and tell a customer that his output stage is discontinued. That was always fun, but so was using a mark IV part where a mark II used to be. They would generally oscillate, that's if the number of pins were the same on each version.
I am so glad those days are almost over. This solution is now called " a chip amp". 😀
-Chris
For the record, I may have been the first to use National chip amps in audio. I used the LM12 as amplifiers for the Entec subwoofer. In the thousands we used only 2 or 3 ever failed.
EUVL said:
I'm so glad to be wrong in this case...but I could swear that I read "OBSOLETE" written over datasheet.
>Chipamp like this?
yep
someways Sakura San
brought legitimicy to monolythyc PAs
Sanyo awaits such an approach
yep
someways Sakura San
brought legitimicy to monolythyc PAs
Sanyo awaits such an approach
From the Zout and PSRR graphs it looks like this opamp might have an OLG corner (transimpedance) below 10kHz...
aparatusonitus said:
No, its the AD846 that is obsolete, the AD844 is in full production and will not be obsoleted in the near future.
You should try it: 2k from pin 2 to gnd, 10k from pin5 to gnd, input pin3. This is a gain of 5 amp, no global feedback, still quite linear with very low distortion. No IM due to feedback-components! Ol bw = cl bw = above 1MHz!
I agree with Dimitri that you can even improve this by an external buffer (from pin 5) instead of the internal buffer at pin 6.
Jan Didden
janneman said:
You are right Jan, it was the AD846
janneman said:
I think I've 2 pcs of AD844 left somewhere, so I might try them as per your suggestion as headphone amp with Allison type buffer, especially 'cos I've discover recently that non global feedback SS amp sounds extraordinary. Thank for the tip Jan!
aparatusonitus said:
One interesting option is to use a pot for the resistor at pin5. That attenuates also the noise ahead of that point if you turn down the volume, making the S/N ratio independent of level.
OTOH a pot before the 844 does lower the distortion when you turn down the volume.
It's a tradeoff, and depends on the line input level to the 844. If it is lower than a volt or so, the pot at pin5 may sound better.
jd
Hi John
“Tom, get the open loop bandwith to 20KHz and THEN we can talk. ”
It’s funny, you know what they say about assuming haha.
I had assumed that with the advances in switching power systems and ever-higher RF goodies, that they would have been there a while ago.
Unexpectedly, well not so much now that I think about why, it’s sort of like development on analogue chips slowed considerably since the late 80’s and early90’s.
Many of the chips in my blue “national semi” data books don’t exist now too.
Conversely, if a wide open loop bw in one component were “the” thing, then we could go back to good tubes like a 6922.
Keeping in mind my experience is semi obsolete, I would still bring up what you see doing the backwards testing I mentioned in my other post.
Here, the open loop bandwidth dictates how low the circuit’s output impedance (sans any series R’s etc) is at any frequency, for an audio amplifier this would be damping factor.
With loudspeakers, one has transducers which nearly always have flaws which increase at a rate greater than the loudness, hence the louder they are, the worse they are.
Here, headroom is the only way to insure a desired level of linearity.
To the degree amplifiers behave in a level sensitive way, then it makes sense to measure them at different levels.
Where I have heard differences in amplifiers in blind testing has been on the decay side of musical transients, not on the loud side.
Where open loop bw corner suggests where the full gain can be used to correct distortion, one might argue that at very low signal levels, the amount of nonlinearity at say “crossover” in a class AB or B may disproportionately “use up” that gain available.
Again, at least with the power amps I was playing with, there was no shortage of odd things on an amplifier’s output terminals when it’s input was grounded and driven though 10 Ohms and another amplifier.
A square wave was interesting too, you could see how long it took to “fight” the current offset on the output.
It seems to me that part of the issue may be that that the operating temperature has a strong effect on some semiconductor properties. In systems which are not class A, the internal temperatures would go up and down with signal history, leaving room for a possible shift in operation having some modestly short time constant.
Loudspeakers have this effect both in their suspension hysteresis and through Voice coil heating.
Anyway, a few thoughts.
Best,
Tom Danley
Danley Sound Labs
“Tom, get the open loop bandwith to 20KHz and THEN we can talk. ”
It’s funny, you know what they say about assuming haha.
I had assumed that with the advances in switching power systems and ever-higher RF goodies, that they would have been there a while ago.
Unexpectedly, well not so much now that I think about why, it’s sort of like development on analogue chips slowed considerably since the late 80’s and early90’s.
Many of the chips in my blue “national semi” data books don’t exist now too.
Conversely, if a wide open loop bw in one component were “the” thing, then we could go back to good tubes like a 6922.
Keeping in mind my experience is semi obsolete, I would still bring up what you see doing the backwards testing I mentioned in my other post.
Here, the open loop bandwidth dictates how low the circuit’s output impedance (sans any series R’s etc) is at any frequency, for an audio amplifier this would be damping factor.
With loudspeakers, one has transducers which nearly always have flaws which increase at a rate greater than the loudness, hence the louder they are, the worse they are.
Here, headroom is the only way to insure a desired level of linearity.
To the degree amplifiers behave in a level sensitive way, then it makes sense to measure them at different levels.
Where I have heard differences in amplifiers in blind testing has been on the decay side of musical transients, not on the loud side.
Where open loop bw corner suggests where the full gain can be used to correct distortion, one might argue that at very low signal levels, the amount of nonlinearity at say “crossover” in a class AB or B may disproportionately “use up” that gain available.
Again, at least with the power amps I was playing with, there was no shortage of odd things on an amplifier’s output terminals when it’s input was grounded and driven though 10 Ohms and another amplifier.
A square wave was interesting too, you could see how long it took to “fight” the current offset on the output.
It seems to me that part of the issue may be that that the operating temperature has a strong effect on some semiconductor properties. In systems which are not class A, the internal temperatures would go up and down with signal history, leaving room for a possible shift in operation having some modestly short time constant.
Loudspeakers have this effect both in their suspension hysteresis and through Voice coil heating.
Anyway, a few thoughts.
Best,
Tom Danley
Danley Sound Labs
janneman said:
Jan did you ever try two with the resistor between the pin2's as an OL diff-out/phase-splitter? Maybe a good front end for bridged amps or circlotrons.
Jan,
What range of close loop gain have you tried with the AD844, and how does that affects the sonics ?
I also find the application in phono interesting :
http://www.lcaudio.com/images/Accessories/riaadia.gif
Patrick
What range of close loop gain have you tried with the AD844, and how does that affects the sonics ?
I also find the application in phono interesting :
http://www.lcaudio.com/images/Accessories/riaadia.gif
Patrick
- Status
- Not open for further replies.