Anybody tried using this auto-biasing chip for a power Mosfet output stage? It controls the currents in the top and bottom Mosfets so that their current product is a constant. Elliminates the need for a bias adjustment and automatically controls crossover for minimum distortion. The technical press reviews were pretty good a few years ago when it first came out, but I haven't seen any published power amplifier designs using it. I purchased a few LT1166 chips recently to try out in a design, but haven't gotten around to it yet. The app. notes for it show it used in an Op. Amp. controlled power amp., but it appears to be perfect for a fully discrete amplifier too. Won't work with lateral mosfets easily though due to operational voltage requirements. But the usual hexfets or vertical mosfets are cheaper and higher current anyway.
I haven't tried these myself, but I have read of others measuring some high frequency distortion when using the LT1166.
I have used the output error correction scheme published by Hawksford in JAES some years ago with good success.
I'd be very curious to hear your opinion if you give the LT part a try.
Phil
I have used the output error correction scheme published by Hawksford in JAES some years ago with good success.
I'd be very curious to hear your opinion if you give the LT part a try.
Phil
amp design
Thanks for the tip! I am thinking of using the chip in an amp. design using an output linearization scheme by Steve Gunderson in JAES Vol 32 No. 6 1984 June pg 430. This approach uses a cascode stage in the voltage amp. stage to provide a connection point for the compensation capacitor which is linearly related to the output rather than to the usual output base or gate drive signal. The cascode's base is slaved to the output. Another interesting variation is apparent if one looks at Seals Optimos Amp. which slaves the cascode to a ratio of the output signal but doesn't impliment the Gunderson compensation pickoff point. Gundersons idea's main attraction is a claimed reduction in high frequency distortion where feedback is least effective, so will be interesting to see if the aforementioned high frequency distortion is cleaned up or not when using the LT1166 chip too.
Thanks for the tip! I am thinking of using the chip in an amp. design using an output linearization scheme by Steve Gunderson in JAES Vol 32 No. 6 1984 June pg 430. This approach uses a cascode stage in the voltage amp. stage to provide a connection point for the compensation capacitor which is linearly related to the output rather than to the usual output base or gate drive signal. The cascode's base is slaved to the output. Another interesting variation is apparent if one looks at Seals Optimos Amp. which slaves the cascode to a ratio of the output signal but doesn't impliment the Gunderson compensation pickoff point. Gundersons idea's main attraction is a claimed reduction in high frequency distortion where feedback is least effective, so will be interesting to see if the aforementioned high frequency distortion is cleaned up or not when using the LT1166 chip too.
Both the schematics for this and the Hawksford circuit
bear a lot of similarity to the original dynamic bias patent
# 3,996,228 from 1976.
I have to say from looking at the data sheets for the LT1166
that I wouldn't be inclined to use it based on performance,
which doesn't seem to be all that great.
bear a lot of similarity to the original dynamic bias patent
# 3,996,228 from 1976.
I have to say from looking at the data sheets for the LT1166
that I wouldn't be inclined to use it based on performance,
which doesn't seem to be all that great.
Don
Thanks for the additional info. I have a copy of that Gunderson article somewhere and will have to look it again in a different light. I'm going to be building an amp using a topology that I think is similar to Gunderson's example (if my memory is right). That amp won't be for music but for a synthesized AC power supply, but the experiment would be fun and educational.
Nelson
Your right! I had seen your patent before (at least the first page from your website) but never noticed the similarities or even thought to compare the different schemes until you pointed this out. Is a complete version of that patent available? And please let me add my thanks to all the others for your new DIY website and all the information you have published over the years for DIYers.
Phil
Thanks for the additional info. I have a copy of that Gunderson article somewhere and will have to look it again in a different light. I'm going to be building an amp using a topology that I think is similar to Gunderson's example (if my memory is right). That amp won't be for music but for a synthesized AC power supply, but the experiment would be fun and educational.
Nelson
Your right! I had seen your patent before (at least the first page from your website) but never noticed the similarities or even thought to compare the different schemes until you pointed this out. Is a complete version of that patent available? And please let me add my thanks to all the others for your new DIY website and all the information you have published over the years for DIYers.
Phil
more than 8 years, this IC is still not a popularity - why (or am I wrong)?
But perhaps there are successors with much more performance - who knows more?
http://www.diyaudio.com/forums/chip-amps/21905-mosfet-audio-driver-2.html
djk
diyAudio Member
Join Date: Feb 2001
Location: USA Quote:
Originally posted by peranders
Linear had one LT1166 but according to reports it wasn't so good. I think it's still best to use discrete driver transistors.
I think the main problem was using pin 2 as the input. If you just drop it in something like a ESP3A in place of the Vbe (bias) transistor you should be fine. The Vas current should be no more than 15mA for this IC to work OK.
*********************
4th April 2001, 01:34 AM #3
djk
diyAudio Member
Join Date: Feb 2001
Location: USA The Linear Technology LT1166 power output stage automatic bias system is what is needed.Do not use the pin 2 mode of operation.Also supports short circuit protection if desired.While designed for FETs it is also good for triple darlington BJT output stages.
*******************
24th January 2005, 05:55 AM #48
djk
diyAudio Member
Join Date: Feb 2001
Location: USA The LT1166 is made for Hexfets, to use with BJTs you need to add a couple of base diodes to increase the voltage drop across the IC (or use a tripple as Jennice suggests). Let pin 2 float and drive both the top and bottom pins instead.
The part is way too much money, and there is no second source.
djk
diyAudio Member
Join Date: Feb 2001
Location: USA Quote:
Originally posted by peranders
Linear had one LT1166 but according to reports it wasn't so good. I think it's still best to use discrete driver transistors.
I think the main problem was using pin 2 as the input. If you just drop it in something like a ESP3A in place of the Vbe (bias) transistor you should be fine. The Vas current should be no more than 15mA for this IC to work OK.
*********************
4th April 2001, 01:34 AM #3
djk
diyAudio Member
Join Date: Feb 2001
Location: USA The Linear Technology LT1166 power output stage automatic bias system is what is needed.Do not use the pin 2 mode of operation.Also supports short circuit protection if desired.While designed for FETs it is also good for triple darlington BJT output stages.
*******************
24th January 2005, 05:55 AM #48
djk
diyAudio Member
Join Date: Feb 2001
Location: USA The LT1166 is made for Hexfets, to use with BJTs you need to add a couple of base diodes to increase the voltage drop across the IC (or use a tripple as Jennice suggests). Let pin 2 float and drive both the top and bottom pins instead.
The part is way too much money, and there is no second source.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.