TDA7052 Replacement Considerations
bradharper
Posts: 64
My ongoing struggles with the TDA7052A(T), while certainly self-induced, have me now officially looking to move to another audio amplifier.
Working through my most recent issue ( discussed @ http://forums.parallax.com/showthread.php?130567-TDA7052-Troubleshooting-Advice-Choppy-Output ) helped me realize that the surface mount version of this chip is designed for a 16ohm speaker, and my project is pretty tightly coupled on an 8 ohm 1W speaker. I'd hoped to work around this issue by lowering the gain using the DC volume control of the chip, and despite two weeks of functioning properly, the audio on my current prototype began a steady decrease in volume today and within 15 minutes diminished to complete silence. There may very well be some other issue in my design which lead to this failure, but I suspect the chip simply cannot handle 8 ohm output.
I've considered the move to another audio amp and identified a potential candidate in the National LM4890 ( http://cache.national.com/ds/LM/LM4890.pdf ). This chip is spec'd appropriately for my needs, is of a similar size and footprint, and has virtually the same part count as the TDA.
One downside that immediately stands out is the supply voltage limit of 6V. One of the original motivations for my choosing the TDA7052 was that its broad supply voltage acceptance allowed me to forgo implementing a second power rail. I have a single switcher (LMZ12001) managing the 3.3V for the propeller and was previously sending all 6-12V into the TDA. If I move to the LM4890, I'll have to establish that 5 or 6V rail, which is concerning because I have very little board space to work with.
Again, my power supply is 6-12V.
Here are my questions related to this potential transition:
1) Any experience with the LM4890? Any other suggestions for my requirements?
2) If I do end up establishing a second power rail, what might be the most practical way to do it in terms of minimizing board space, maximum efficiency. My board will be enclosed with little airflow to dissipate heat, so I do have to be concerned with ambient temps and the thermal considerations of whatever form of regulation I go with.
3) Would the switcher be more appropriate for the 5V rail with some other convention handle the 3.3?
4) Is this move even necessary, or is there a way I can make the TDA work well within my needs?
In summary, moving to another amp likely means a shift in power management strategy, more parts, more heat, etc. - from experience, does this look like a necessary hurdle? As always, I greatly appreciate everyone's time and perspective.
Working through my most recent issue ( discussed @ http://forums.parallax.com/showthread.php?130567-TDA7052-Troubleshooting-Advice-Choppy-Output ) helped me realize that the surface mount version of this chip is designed for a 16ohm speaker, and my project is pretty tightly coupled on an 8 ohm 1W speaker. I'd hoped to work around this issue by lowering the gain using the DC volume control of the chip, and despite two weeks of functioning properly, the audio on my current prototype began a steady decrease in volume today and within 15 minutes diminished to complete silence. There may very well be some other issue in my design which lead to this failure, but I suspect the chip simply cannot handle 8 ohm output.
I've considered the move to another audio amp and identified a potential candidate in the National LM4890 ( http://cache.national.com/ds/LM/LM4890.pdf ). This chip is spec'd appropriately for my needs, is of a similar size and footprint, and has virtually the same part count as the TDA.
One downside that immediately stands out is the supply voltage limit of 6V. One of the original motivations for my choosing the TDA7052 was that its broad supply voltage acceptance allowed me to forgo implementing a second power rail. I have a single switcher (LMZ12001) managing the 3.3V for the propeller and was previously sending all 6-12V into the TDA. If I move to the LM4890, I'll have to establish that 5 or 6V rail, which is concerning because I have very little board space to work with.
Again, my power supply is 6-12V.
Here are my questions related to this potential transition:
1) Any experience with the LM4890? Any other suggestions for my requirements?
2) If I do end up establishing a second power rail, what might be the most practical way to do it in terms of minimizing board space, maximum efficiency. My board will be enclosed with little airflow to dissipate heat, so I do have to be concerned with ambient temps and the thermal considerations of whatever form of regulation I go with.
3) Would the switcher be more appropriate for the 5V rail with some other convention handle the 3.3?
4) Is this move even necessary, or is there a way I can make the TDA work well within my needs?
In summary, moving to another amp likely means a shift in power management strategy, more parts, more heat, etc. - from experience, does this look like a necessary hurdle? As always, I greatly appreciate everyone's time and perspective.
Comments
Why do you "suspect" that it has anything to do with the 8Ω? It's rated for it.
Why do you feel that you'll have a different result with another IC?
Isn't that the trouble - suspicions, hunches, and such?
I'm not having a dig at you, just your perspective.
There are lots of audio ICs out there.
What's so horrible about an LM386?
Brad, I see the LM4890 is good for 400mW into 8Ω at 3.3V supply, would that be sufficient power?
Also, as PJ mentioned the LM386 may be an option, it is available in SOIC and works over your 6V to 12V range, it needs a large capacitor on the output, but that may be less space than a seperate 5V regulator.
C.W.
BTW, the S2 robot uses an LM4990.
-Phil
As far as the LM386, I have tested with that chip some, but moved to the TDA based on recommendations here and elsewhere, and due in part to my lack of board space in relation to the parts it needs.
@ctwardell - I very well might have misread the sheet, but under the 5V power ratings it lists:
Po Output Power (8Ω) THD = 2% (max); f = 1 kHz 1.0 0.8 W
Am I looking at the correct spec?
@ Pliers - Hmm... I'd considered that, but had yet to try it out. I'd still prefer to avoid a major revision at this stage and just stay with the TDA. So long as I can maintain reasonable volume and clarity I'm open to whatever will make it work.
As far as the 8Ω resistor in series -- No; it's about impedance, not resistance (go back to school, or start)
"Yesterday's news" -- Oh, dear, Mr. Blackwell, excuse the heck out of me.
See, bradharper, it's not about what works, it's about a sexy IC with just the right accessories - Ask Phil.
Well, here's a little "fashion tip": they make tantalum caps in pretty small sizes these days.
Yeah, I think those are the correct ratings @5V, what I was getting at was can you get by at 400mW and just power the LM4890 from your existing 3.3V supply.
Ah.. gotcha. Hmm.. I definitely should consider that. Thx!
I did a little searching (google) to find out why a resistor would not work, did not find an answer.
Never left school, still learning.
I did a little searching (google) to find out why a resistor would not work, did not find an answer."
Neither multi-kilowatt lasers nor automation systems have I... it's basic electronics.
I'll cast a few pearls:
An impedance is not the same as a resistance.
Look into AC circuit analysis, inductive reactance (and capacitive reactance.)