On Thursday, February 16, 2023 at 4:54:24 PM UTC-5, John Walliker wrote:> On Thursday, 16 February 2023 at 18:08:05 UTC, Ricky wrote: > > On Thursday, February 16, 2023 at 9:04:24 AM UTC-5, piglet wrote: > > > On 16/02/2023 08:26, Ricky wrote: > > > > On Wednesday, February 15, 2023 at 5:47:32 PM UTC-4, Lasse Langwadt Christensen wrote: > > > >> onsdag den 15. februar 2023 kl. 06.47.09 UTC+1 skrev Ricky: > > > >>> I have an audio amplifier circuit that uses an op amp I can't buy anymore because of the semiconductor shortage. It's not extremely special, but it drives highly capacitive loads (like a few hundred feet of cable) and will drive a 62.1 ohm load with 75 mA. I'd actually like a higher drive, but this was the best I could find in 2008 when I designed it and I've found nothing better since. > > > >>> > > > >>> Two main requirements are low power consumption (the LM8272 is around 1 mA per amp) and small size (8 pin MSOP dual). > > > >>> > > > >>> I'm wondering if I might replace these with a push-pull transistor output stage. The outputs are differential, so I would need an inverter to drive them. The current circuit uses positive feedback to allow a 12.1 ohm resistor to emulate a 50 ohm output impedance. In the new design this will have an analog switch to change between 300 ohm and 50 ohm. I can't see a useful way to do this same trick with a push-pull output. So this idea is not looking so great. > > > >>> > > > >>> I looked at audio drivers for headphones and speakers. The lower power devices all run on 5V power, while I need 12V power. The devices that use 12V power are pretty much all class D or similar and require output filtering which uses too much space. > > > >>> > > > >>> I've even looked for amps with differential outputs, but nothing workable. Price is not a huge issue, but some of these parts are $10 a pop and I need two channels of differential drivers. $40 is a bit steep. > > > >>> > > > >>> Any ideas? > > > >> I don't see how you can possible get anywhere near as small as what you have unless you find another dual opamp > > > > > > > > The two channels (think stereo) of differential output would map well to audio circuits, but I can't find a good chip to work with a 12V supply. > > > > > > > > > > > >> how about https://www.digikey.dk/en/products/detail/analog-devices-inc/AD8012ARMZ/620444 ? > > > > > > > > Looks good in some respects, but it is also not rated for 12V power. At ±5V it will swing ±4V, but that I can get from other parts... I think. Most of them are not well characterize in this regard. For whatever reason, they often list a minimum short circuit current. > > > > > > > > The absolute maximum power is 12.6V, so maybe it would run from 12V power. Funny, they didn't spec it, if it is intended for that. > > > > > > > > Thanks for the part number. > > > > > > > L272 and TCA0372 are well known dual power op amps. Maybe not HiFi audio > > > but should be ok for telecoms audio. > > Thanks. The L272 has the drive capability for sure, but it's in far too large a package. Same for the TCA0372. These parts seem targeted to motor drive. > Another one that might be worth a look is the ALM2403-Q1. It isn't targeted at audio, but > it does look interesting. > The TPA6120A2DWP is a headphone driver which might do what you want. However, > the data sheet doesn't make it easy to be sure. > > JohnThe ALM is too large at twice the size of the current part. The TPA6120A2RGYR has possibilities. The QFN package is smaller than the MSOP8 and the part is available. Arrow seems to have 9,000 in stock, but they might be reflecting the 9,000 that TI is stocking. Either way, it's at least 9,000 parts and Digikey is saying they will have more in six weeks. That's the best I've seen so far. I just wish it were not in a package that makes it a sole sourced part. I made 8,000 of these boards last year and could not get all the LM8272 parts I needed. I got some at Winsource, but I was able to fill in using compatible parts for two sockets, that didn't need to drive the outputs, so much lower current requirement. With these parts, there's not likely to be a suitable alternate. Still, better to be able to make something, than nothing. Thanks. -- Rick C. +- Get 1,000 miles of free Supercharging +- Tesla referral code - https://ts.la/richard11209
High Current Drive Output Stage
Started by ●February 15, 2023
Reply by ●February 16, 20232023-02-16
Reply by ●February 18, 20232023-02-18
On 2/16/2023 6:53 PM, Ricky wrote:> On Thursday, February 16, 2023 at 4:54:24 PM UTC-5, John Walliker wrote: >> On Thursday, 16 February 2023 at 18:08:05 UTC, Ricky wrote: >>> On Thursday, February 16, 2023 at 9:04:24 AM UTC-5, piglet wrote: >>>> On 16/02/2023 08:26, Ricky wrote: >>>>> On Wednesday, February 15, 2023 at 5:47:32 PM UTC-4, Lasse Langwadt Christensen wrote: >>>>>> onsdag den 15. februar 2023 kl. 06.47.09 UTC+1 skrev Ricky: >>>>>>> I have an audio amplifier circuit that uses an op amp I can't buy anymore because of the semiconductor shortage. It's not extremely special, but it drives highly capacitive loads (like a few hundred feet of cable) and will drive a 62.1 ohm load with 75 mA. I'd actually like a higher drive, but this was the best I could find in 2008 when I designed it and I've found nothing better since. >>>>>>> >>>>>>> Two main requirements are low power consumption (the LM8272 is around 1 mA per amp) and small size (8 pin MSOP dual). >>>>>>> >>>>>>> I'm wondering if I might replace these with a push-pull transistor output stage. The outputs are differential, so I would need an inverter to drive them. The current circuit uses positive feedback to allow a 12.1 ohm resistor to emulate a 50 ohm output impedance. In the new design this will have an analog switch to change between 300 ohm and 50 ohm. I can't see a useful way to do this same trick with a push-pull output. So this idea is not looking so great. >>>>>>> >>>>>>> I looked at audio drivers for headphones and speakers. The lower power devices all run on 5V power, while I need 12V power. The devices that use 12V power are pretty much all class D or similar and require output filtering which uses too much space. >>>>>>> >>>>>>> I've even looked for amps with differential outputs, but nothing workable. Price is not a huge issue, but some of these parts are $10 a pop and I need two channels of differential drivers. $40 is a bit steep. >>>>>>> >>>>>>> Any ideas? >>>>>> I don't see how you can possible get anywhere near as small as what you have unless you find another dual opamp >>>>> >>>>> The two channels (think stereo) of differential output would map well to audio circuits, but I can't find a good chip to work with a 12V supply. >>>>> >>>>> >>>>>> how about https://www.digikey.dk/en/products/detail/analog-devices-inc/AD8012ARMZ/620444 ? >>>>> >>>>> Looks good in some respects, but it is also not rated for 12V power. At ±5V it will swing ±4V, but that I can get from other parts... I think. Most of them are not well characterize in this regard. For whatever reason, they often list a minimum short circuit current. >>>>> >>>>> The absolute maximum power is 12.6V, so maybe it would run from 12V power. Funny, they didn't spec it, if it is intended for that. >>>>> >>>>> Thanks for the part number. >>>>> >>>> L272 and TCA0372 are well known dual power op amps. Maybe not HiFi audio >>>> but should be ok for telecoms audio. >>> Thanks. The L272 has the drive capability for sure, but it's in far too large a package. Same for the TCA0372. These parts seem targeted to motor drive. >> Another one that might be worth a look is the ALM2403-Q1. It isn't targeted at audio, but >> it does look interesting. >> The TPA6120A2DWP is a headphone driver which might do what you want. However, >> the data sheet doesn't make it easy to be sure. >> >> John > > The ALM is too large at twice the size of the current part. The TPA6120A2RGYR has possibilities. The QFN package is smaller than the MSOP8 and the part is available. Arrow seems to have 9,000 in stock, but they might be reflecting the 9,000 that TI is stocking. Either way, it's at least 9,000 parts and Digikey is saying they will have more in six weeks. That's the best I've seen so far. I just wish it were not in a package that makes it a sole sourced part. > > I made 8,000 of these boards last year and could not get all the LM8272 parts I needed. I got some at Winsource, but I was able to fill in using compatible parts for two sockets, that didn't need to drive the outputs, so much lower current requirement. With these parts, there's not likely to be a suitable alternate. > > Still, better to be able to make something, than nothing. Thanks. >Cold these be any use? <https://www.lcsc.com/product-detail/Audio-Power-OpAmps_UTC-Unisonic-Tech-MC34119G-SO8-R_C87284.html> They are discontinued by NXP and but looks like someone in China is still making 'em. The normal quiescent current is a little high at 3.3 mA typ @ 16 V but it does have a power down pin to bring it well below 1 mA.
Reply by ●February 18, 20232023-02-18
On Saturday, February 18, 2023 at 11:04:04 AM UTC-5, bitrex wrote:> On 2/16/2023 6:53 PM, Ricky wrote: > > On Thursday, February 16, 2023 at 4:54:24 PM UTC-5, John Walliker wrote: > >> On Thursday, 16 February 2023 at 18:08:05 UTC, Ricky wrote: > >>> On Thursday, February 16, 2023 at 9:04:24 AM UTC-5, piglet wrote: > >>>> On 16/02/2023 08:26, Ricky wrote: > >>>>> On Wednesday, February 15, 2023 at 5:47:32 PM UTC-4, Lasse Langwadt Christensen wrote: > >>>>>> onsdag den 15. februar 2023 kl. 06.47.09 UTC+1 skrev Ricky: > >>>>>>> I have an audio amplifier circuit that uses an op amp I can't buy anymore because of the semiconductor shortage. It's not extremely special, but it drives highly capacitive loads (like a few hundred feet of cable) and will drive a 62.1 ohm load with 75 mA. I'd actually like a higher drive, but this was the best I could find in 2008 when I designed it and I've found nothing better since. > >>>>>>> > >>>>>>> Two main requirements are low power consumption (the LM8272 is around 1 mA per amp) and small size (8 pin MSOP dual). > >>>>>>> > >>>>>>> I'm wondering if I might replace these with a push-pull transistor output stage. The outputs are differential, so I would need an inverter to drive them. The current circuit uses positive feedback to allow a 12.1 ohm resistor to emulate a 50 ohm output impedance. In the new design this will have an analog switch to change between 300 ohm and 50 ohm. I can't see a useful way to do this same trick with a push-pull output. So this idea is not looking so great. > >>>>>>> > >>>>>>> I looked at audio drivers for headphones and speakers. The lower power devices all run on 5V power, while I need 12V power. The devices that use 12V power are pretty much all class D or similar and require output filtering which uses too much space. > >>>>>>> > >>>>>>> I've even looked for amps with differential outputs, but nothing workable. Price is not a huge issue, but some of these parts are $10 a pop and I need two channels of differential drivers. $40 is a bit steep. > >>>>>>> > >>>>>>> Any ideas? > >>>>>> I don't see how you can possible get anywhere near as small as what you have unless you find another dual opamp > >>>>> > >>>>> The two channels (think stereo) of differential output would map well to audio circuits, but I can't find a good chip to work with a 12V supply. > >>>>> > >>>>> > >>>>>> how about https://www.digikey.dk/en/products/detail/analog-devices-inc/AD8012ARMZ/620444 ? > >>>>> > >>>>> Looks good in some respects, but it is also not rated for 12V power. At ±5V it will swing ±4V, but that I can get from other parts... I think. Most of them are not well characterize in this regard. For whatever reason, they often list a minimum short circuit current. > >>>>> > >>>>> The absolute maximum power is 12.6V, so maybe it would run from 12V power. Funny, they didn't spec it, if it is intended for that. > >>>>> > >>>>> Thanks for the part number. > >>>>> > >>>> L272 and TCA0372 are well known dual power op amps. Maybe not HiFi audio > >>>> but should be ok for telecoms audio. > >>> Thanks. The L272 has the drive capability for sure, but it's in far too large a package. Same for the TCA0372. These parts seem targeted to motor drive. > >> Another one that might be worth a look is the ALM2403-Q1. It isn't targeted at audio, but > >> it does look interesting. > >> The TPA6120A2DWP is a headphone driver which might do what you want. However, > >> the data sheet doesn't make it easy to be sure. > >> > >> John > > > > The ALM is too large at twice the size of the current part. The TPA6120A2RGYR has possibilities. The QFN package is smaller than the MSOP8 and the part is available. Arrow seems to have 9,000 in stock, but they might be reflecting the 9,000 that TI is stocking. Either way, it's at least 9,000 parts and Digikey is saying they will have more in six weeks. That's the best I've seen so far. I just wish it were not in a package that makes it a sole sourced part. > > > > I made 8,000 of these boards last year and could not get all the LM8272 parts I needed. I got some at Winsource, but I was able to fill in using compatible parts for two sockets, that didn't need to drive the outputs, so much lower current requirement. With these parts, there's not likely to be a suitable alternate. > > > > Still, better to be able to make something, than nothing. Thanks. > > > Cold these be any use? > > <https://www.lcsc.com/product-detail/Audio-Power-OpAmps_UTC-Unisonic-Tech-MC34119G-SO8-R_C87284.html> > > They are discontinued by NXP and but looks like someone in China is > still making 'em. > > The normal quiescent current is a little high at 3.3 mA typ @ 16 V but > it does have a power down pin to bring it well below 1 mA.I think the specs are ok, but I'm not sure I can configure it for the synthetic output impedance circuit I'm using now. But even if I figure that out, there's the problem of supply. I'm not designing in a part that is only available through Chinese suppliers. Actually, I see Avnet has them in stock in the TSSOP package. The other packages are too large. Even the TSSOP is bigger than the MSOP, but not by a lot. I'll look at this part a bit harder. Thanks. -- Rick C. ++ Get 1,000 miles of free Supercharging ++ Tesla referral code - https://ts.la/richard11209
Reply by ●February 21, 20232023-02-21
I think I found a viable replacement. The LM8272 is a TI part. They make an LM8262 which is very similar, with a lower current drive. I have not had much luck using their search pages, because they split their op amps into different groups, and each one has to be searched separately. But today I looked in General Amps again and found several parts that are very similar, with one having the same output drive level, LM7332. 12,000 in stock! I need to order a few parts and see if they actually work the same under the loading conditions. -- Rick C. --- Get 1,000 miles of free Supercharging --- Tesla referral code - https://ts.la/richard11209
