I am looking for guidance to utilize PA series chips. Was curious to see if anyone has had experience with their products.
thanks
hoibit
thanks
hoibit
I have used Apex power opamps, though not for audio. But some of their designs seem to be very suitable, ie the PA05. Support and documentation is as excellent as the product are. But I fear that the designs that would make a killer "chip amp" like the PA05 are beyond the average DIYer's level in terms of cost (the PA05 was above $300 last time I checked, IIRC).hoibit said:
- Klaus
The PA19 is great and I suspect Apex's best sounding chip with moderate power. The PA05 is less transparent, though still very good, and more powerful. I used to listen to a 250 watt Krell at work and my memory was that the PA05 was in the same class. Unfortunately, my comparisons were not scientific.
I also own the PA45 as a affordable audio amp. It is pretty good up to 2 khz. But its high frequencies are mushy. I used it successfully as a woofer amp in a triamp project.
Caution: Rant following!
I avoided the recent debate which asked if any amps have a different sound. But an excellent example is the PA05 and the PA45. The PA05 is sold as a possible audiophile amp, the PA45 is sold as a possible PA amp. The PA05 has lower harmonic distortion than the PA45. And guess what, the PA05 sounds better than the PA45. These two amps are marketed different, measure different, and they sound different. QED There is an audible difference between amps.
Rant over.
Still, I like the PA45 in the low frequencies. I would love to know the harmonic distortion of the PA 19. It isn't include in the data sheet. I suspect they had trouble measuring it.
BTW: Apex has great application notes and white papers dealing with all aspects of amplifier application.
At long last the conclusion:
If you have the money and don't need a lot of power go with the PA19. If you have the money and need the power, go with the PA05. If you don't want to spend a lot of money and you can deal with a more involved application: take a look at the new National Semiconductor input stage chips, particularly LME49810 in general or the LME49830 for mosfet outputs in particular. (I built some demo amps for National using the LM4702, which sound great, but the chips have limited output current to drive the output stage and are tricky to avoid noise.) I will be trying the LME49810 in the future.
After thoughts: The Apex amps are easy to apply and that is definitely worth some up front cost. However, I was working on a PA05 and something shorted, and the chip fried. This my only "failure" of an Apex chip (likely operator error). But it was financially painful. My lesson is to work carefully. Second thought, I suspect that a good application of the LME##### National chips could be better than the PA19. At this level, it will be very difficult to hear differences. But a beginner would be better off with the Apex chips. Alternately, beginner might want to work with the LM3886 first. It is similar to the PA45 but cheaper. It is not as good as the PA45 on the low end, but may be a bit better on the high frequency end (based on old memories).
I also own the PA45 as a affordable audio amp. It is pretty good up to 2 khz. But its high frequencies are mushy. I used it successfully as a woofer amp in a triamp project.
Caution: Rant following!
I avoided the recent debate which asked if any amps have a different sound. But an excellent example is the PA05 and the PA45. The PA05 is sold as a possible audiophile amp, the PA45 is sold as a possible PA amp. The PA05 has lower harmonic distortion than the PA45. And guess what, the PA05 sounds better than the PA45. These two amps are marketed different, measure different, and they sound different. QED There is an audible difference between amps.
Rant over.
Still, I like the PA45 in the low frequencies. I would love to know the harmonic distortion of the PA 19. It isn't include in the data sheet. I suspect they had trouble measuring it.
BTW: Apex has great application notes and white papers dealing with all aspects of amplifier application.
At long last the conclusion:
If you have the money and don't need a lot of power go with the PA19. If you have the money and need the power, go with the PA05. If you don't want to spend a lot of money and you can deal with a more involved application: take a look at the new National Semiconductor input stage chips, particularly LME49810 in general or the LME49830 for mosfet outputs in particular. (I built some demo amps for National using the LM4702, which sound great, but the chips have limited output current to drive the output stage and are tricky to avoid noise.) I will be trying the LME49810 in the future.
After thoughts: The Apex amps are easy to apply and that is definitely worth some up front cost. However, I was working on a PA05 and something shorted, and the chip fried. This my only "failure" of an Apex chip (likely operator error). But it was financially painful. My lesson is to work carefully. Second thought, I suspect that a good application of the LME##### National chips could be better than the PA19. At this level, it will be very difficult to hear differences. But a beginner would be better off with the Apex chips. Alternately, beginner might want to work with the LM3886 first. It is similar to the PA45 but cheaper. It is not as good as the PA45 on the low end, but may be a bit better on the high frequency end (based on old memories).
thanks for the help.
Benchtester,
i had a student do a similar 'test' of what would have been the sixth pa04 but now i have 5 PA04's available. i do have a copy of the latest book that Apex publishes for their product line including the app notes. the issues that i am attempting to jump over are small but numerous. mostly to do with the mass of information that they try to throw at the reader. nothing really major, just enough to make me want to ask what others have accomplished in the past. and if there are any examples or schematics to peruse and ponder.
so, basically what i'm thinking is push-pull and use the 5th one as a subwoofer driver.
i also have access to a lpkf circuit board plotter and the associated cam program. you definitely have to get one of those.
i'm also trying to decide on cooling. i do have the oem heatsinks that have the optional water cooling passages. i have remachined the heatsinks to have hole spacings for two chips for each heatsink and am in the process of making a plate of copper plate with pocket to cover the bare side of the chip and to attach an additional copper cpu heatsink. the aim being very quiet yet with the ability to progressively add cooling fluid flow as needed to maintain a stable temperature.
but, let me of course state the obvious that i would ask if i was presented with such statements.
i do not listen at unrealistic levels. i twinge at the systems that seem to have pervaded the asphalt lately.
it would be nice to be able to drowned them out though.
so, long post and short ending question. are there any pitfalls that i
should consider that you may have already run into regarding protection caps, feedback, etc?
thanks much and wish me luck.
Benchtester,
i had a student do a similar 'test' of what would have been the sixth pa04 but now i have 5 PA04's available. i do have a copy of the latest book that Apex publishes for their product line including the app notes. the issues that i am attempting to jump over are small but numerous. mostly to do with the mass of information that they try to throw at the reader. nothing really major, just enough to make me want to ask what others have accomplished in the past. and if there are any examples or schematics to peruse and ponder.
so, basically what i'm thinking is push-pull and use the 5th one as a subwoofer driver.
i also have access to a lpkf circuit board plotter and the associated cam program. you definitely have to get one of those.
i'm also trying to decide on cooling. i do have the oem heatsinks that have the optional water cooling passages. i have remachined the heatsinks to have hole spacings for two chips for each heatsink and am in the process of making a plate of copper plate with pocket to cover the bare side of the chip and to attach an additional copper cpu heatsink. the aim being very quiet yet with the ability to progressively add cooling fluid flow as needed to maintain a stable temperature.
but, let me of course state the obvious that i would ask if i was presented with such statements.
i do not listen at unrealistic levels. i twinge at the systems that seem to have pervaded the asphalt lately.
it would be nice to be able to drowned them out though.
so, long post and short ending question. are there any pitfalls that i
should consider that you may have already run into regarding protection caps, feedback, etc?
thanks much and wish me luck.
Best wishes for you project.
I don't have very much advice since the Apex amps are just like regular op amps except more power.
For the PA19 they suggest avoiding high impedance feedback. For audio amps this can be simplified to using a 500 ohm (or less) on the gain (input) resistor. I only had 10x gain, so I had a 5000 ohm feedback resistor. It appears that this is less of an issue with the PA04, but I think it might be a good starting place.
As you will have read, use bypass caps close (within 1.5 inches) on the power pins. I would recommend film and or tantalum caps closest, and then a 1000uF electrolytic right after it.
You might want to limit the current to 1 amp until you have everything working right and playing music at low level. Then to change to the higher current limit for actual usage.
I can not remember having any troubles with getting the amps to work right.
Good luck and enjoy!
I don't have very much advice since the Apex amps are just like regular op amps except more power.
For the PA19 they suggest avoiding high impedance feedback. For audio amps this can be simplified to using a 500 ohm (or less) on the gain (input) resistor. I only had 10x gain, so I had a 5000 ohm feedback resistor. It appears that this is less of an issue with the PA04, but I think it might be a good starting place.
As you will have read, use bypass caps close (within 1.5 inches) on the power pins. I would recommend film and or tantalum caps closest, and then a 1000uF electrolytic right after it.
You might want to limit the current to 1 amp until you have everything working right and playing music at low level. Then to change to the higher current limit for actual usage.
I can not remember having any troubles with getting the amps to work right.
Good luck and enjoy!
APEX PA05
If anybody needs some of these amps I have 5 new ones. Was going to build an amp, don't have the time now.
Thanks
If anybody needs some of these amps I have 5 new ones. Was going to build an amp, don't have the time now.
Thanks
Just got a few PA19 try to build a headphone amp. Benchtester, could you email me the schematics of your pa19 amp? My email: tenryfu@yahoo.com
Many thanks!
Many thanks!
- Status
- Not open for further replies.