TAS5630 - volume control

Hi, Calum,

We don't have a lot of experience with these types of applications, but in general, they should theoretically work fine (and I have heard several customers being successful).

To answer your questions:

1) The customer needs to attenuate the input signal size according to the level needed to get 24VAC output level. For example, if their input signal is 1VAC, then they would need a gain of 24V/V in order to achieve 24VAC output. The amplifier has a gain of 20 dB (10V/V).

2) Someplace around 35-40V would be the minimum needed. There is no issue using the 48VDC supply, though.

3) This depends on the design of your heat sinking and the ability to get the heat away from our IC. It is a system-level design trade-off.

4) Either one would probably work.

-d2

Don:

Calum posted this for me on the forum because I was having difficulty.  (For some reason Firefox can't find your server.)

Thanks for your response.  You did clear up some of the muddy water.  In terms of choosing between the TAS5630 and the TAS5613, I suppose the bigger question would be which part is stronger from a sales perspective.  I would hate to design in the TAS5613 only to have it discontinued in a few years because there was more of a demand for the other part.  Also, in knowing that my ability to remove heat from the enclosure is restricted, perhaps the TAS5630 would be more up to the challenge, especially when the output is throttled back to 60W per channel.

I thought your answer to question three lacked any real information.  We all know that heat is an issue and the bottom line is the temperature of the IC must remain at an acceptable level to achieve performance.  Let's assume that the assembly under discussion is the TAS5630DKD2EVM with the heat sink provided.  If this demo were to be placed inside a closed system and both amplifiers were running at 60 Watts, would you expect the IC to eventually self-destruct with a continuous duty cycle? (I noticed that the demo used the better part for thermal dissipation.)  If no efforts were made to transfer the heat from the pre-existing heat sink to the enclosure, would we just have to reduce the duty cycle?  It appears that the main objective here is to stay below the maximum thermal junction temperature of 150 degree Centigrade.  Is this a correct assessment?  Would you have any idea how long we could drive the TAS5630DKD2EVM when delivering 25VAC and 60 Watts?

Regards,

Anthony

Hi, Anthony,

TI has a very strict anti-obsolescence policy, so you don't need to worry about this when selecting TI components... I can tell you we sell more of the TAS5613s than 30s, though.

Thermal is very complex topic because our performance is thermal related, so it's an iterative process to try and calculate. We used to use a thermal modeling tool to do this type of work, but unfortunately, we don't have access to that software any longer.

60W should be fine with the EVM and heatsink on it, even in a closed box. The EVM can't do 2x300W even in free air, you have to have some airflow across the heat sink to accomplish that. I am not sure exactly where the break point is, and as I mentioned, we have no real good way to model that... Perhaps you can have Calum get you an EVM and try it out to see if it meets your requirements - I think it should, but I won't guarantee that...

-d2

Don:

Thanks for your response.  It was helpful.

I have been doing a comparison of the TAS5613 and the TAS5630.  Our target for delivered power is 2 x 60W @24 VAC.  Both data sheets indicate that the TAS5613 is more efficient with these output parameters.  It’s confusing to me how the heat dissipation rating is not as good for the TAS5613.  You would expect the case temperature rise to be less for the delivered power.

It would appear that both parts are pin compatible.  It is unclear to me if the TAS5613 is available in the DKD package.  The data sheet I have shows a “Product Preview” for that package, implying that it is not in production.

Regards,

Anthony

Anthony,

What data are you comparing?

As for the DKD package, it is our intention to release it for the TAS5613, but as far as I know, we have no schedule to do such.

So, if that's important to you, I'd go w/the 5630 since we have that one released today. it is a lot more forgiving for the mechanical assembly of the heat sink.

-d2

Don:

I was looking at the efficiency curves as stated on the data sheets for the TAS5613 and the TAS5630.  It is hard for me to believe that these curves are accurate because the figure for degrees C per Watt is less for the TAS5630.  With this in mind you would assume that the TAS5630 is more efficient.  Does this make sense to you or am I missing something?

The amplifier will be running in the neighborhood of 60 Watts.  A heat sink will be required because I will be operating in a closed environment.  My concerns are for efficiency and keeping the heat sink temperature as low as possible.  Your response indicates that a TAS5630 would be preferred, using the DKD package.  Is this a correct assessment?

In another posting I indicated that I may need to provide 70VAC audio along with 25VAC.  There appears to be other concerns with using a step up transformer.  The addition of a transformer would require the amplifier to work even harder and efficiency becomes more of an issue.

Re: TAS5613/TAS5630 - 70 Volt Audo

Anthony

Hi, Anthony,

The difference is supply voltage. The TAS5613 is running at a lower supply voltage which makes it more efficient than the 5630 even though they're running the same output power. If you adjust the supply voltage down on the 5630, you will see a similar plot, I assume.

For this discussion, the main difference between these two parts is the operating voltage. 5630 is rated for a higher voltage, so while they both have the same rds(on) fets, the 5630 die is actually bigger which gives it better thermal dissipation as you have pointed out.

You can't go wrong with the DKD package! Actually, we were having an argument here about that - the DKD is awesome from a mechanical/thermal standpoint, but the PHP offers so many more ground pins that grounding is better. Nothing is ever easy, is it?

I saw Steve commented on your other thread.

-d2