[phaedrus]

I ran out of parts to attach to my PPA board, so I started work on my battery board. This results in a couple of questions:

1. Using the 24V elpac supply, what is the effect of attempting to charge 15 or 18 AAA 800mA batteries? If I jump up to the 30V I'm still shy of what Tangent's calculator shows I need for a full set of 18. Is it a situation where I've got to choose batteries or opamps; or, can I have batteries AND opamp flexibility at the cost of a longer/slower charge time?
2. I'm a touch confused about how to connect power from the wall supply and how to connect the battery board to the amp board. I purchased the recommended molex connectors and pins and I seem to recall something about Y-ing V+ during the battery board thread. I think that I understand the diode bridge, at least I like to think that my head is wrapped most of the way around that...
   
   hmmm --
   
   Writing this and staring hard at the schematics caused me to think perhaps the molex connectors chain off the wall supply jack like the power connectors inside a PC (v+ and v- from the power jack to battery board; v+, b+ and v- from battery board to amp board). Am I in the right ballpark?
   

Thank you for your help in advance.

[edit: corrected me bad grammah]

 

[tangent]

Quote:

Using the 24V elpac supply, what is the effect of attempting to charge 15 or 18 AAA 800mA batteries?

Right on the edge of this condition, the output current of the regulator starts to drop. As the supply voltage drops relative to the total drops in the charging circuit, current continues to drop through the regulator until no charging takes place except through the trickle charge path.

If you're really stuck with the 24V supply and must have more than 12 cells, go with a trickle-only charger and an LDO regulator. This will let you get to 14 cells.

Quote:

If I jump up to the 30V I'm still shy of what Tangent's calculator shows I need for a full set of 18.

That's true. You want 31 or 32V to charge 18 cells. Are you sure you need 18 cells, though?

Quote:

I'm a touch confused about how to connect power from the wall supply and how to connect the battery board to the amp board.

DC input jack to V+ and V- pads on the amp board, V+/V-/B+ from the amp board to a Molex female which connects to a Molex male on the battery board. There are other ways, but that's the one I like.

 

[phaedrus]

Quote:

Originally posted by tangent Are you sure you need 18 cells, though?

I'm pretty sure that 18 is a testosterone want thing rather than a need. My anticipated use cycle will look something like this:

1. Public transit to and from work, 1.5 hours off the grid, twice per day (2-3 days per week, I drive myself in on studio class days)
2. 6-12 hours in the office, on the grid
3. 3-6 hours in studio, not able to easily attach to the grid
4. 6-8 hour stretches in my home studio on Saturday/Sunday where off grid would be most convenient
5. Senn HD600's almost exclusively while connected to the grid -- they rarely leave my office
6. Grado SR325's or Ety ER-4S's for off grid use

I'm happy to leave it plugged in over night with my PCDP, cell phone, pda, iPod, laptop, etc. -- what's one more portable device?

Are 12 AAA's enough? Do I need to switch to 10 AA's? Is the trade off in cell size voltage and usable panel real estate for longer run times? At the moment, I would prefer the flexibility to switch between each of the three sets of opamps (ad8610, opa627, opa637) I have for the PPA -- at least until I am able to make an informed choice.

Writing this introduces another question to my mind -- should one of the battery holders be situated over R7? I infer from what I've read that a thermistor could go there at some point in the future and would need a cell above it to work.

[edit: and by ad8210 I really meant ad8610...]

 

[tangent]

Quote:

I'm pretty sure that 18 is a testosterone want thing

How about this for some counter-testosterone: more cells in series means a higher supply impedance, which means a higher ripple voltage on the rails. Also, more cells means higher heat and thus shorter cell life. And, if you don't reach the clipping point before the cells die with 12 cells, 18 cells won't last any longer and the extra voltage may not give you an audible improvement.

Think of it like choosing an aluminum 4-cylinder engine for your rice rocket instead of trying to cram a cast iron 8 in there. The 4 banger is easier to apply and it'll probably push the car faster, too.

Quote:

Do I need to switch to 10 AA's?

That'll buy you longer run times and still lower supply impedance. (Even lower than 10xAAA, due to the greater cross-section of the cells.) The bigger question is, would you be happy with the consequences in terms of case layout if you go with AA's?

Quote:

opa627, opa637

I don't think I'd run those from just 12V. That's too close to the clipping point with some common heapdhones.

Quote:

should one of the battery holders be situated over R7? I infer from what I've read that a thermistor could go there at some point in the future and would need a cell above it to work.

If you plan on setting up temp-based fast charge termination, definitely put R7 under one of the cells. It wouldn't be a total loss if you didn't do this, but the thermistor wouldn't work as well if it had to wait for the heated air to circulate before it could sense an overtemp condition.

 

[phaedrus]

Quote:

Originally posted by tangent a cast iron 8 in there

but i like my 409...

I think that you are encouraging me to solve this for myself, which is good. I feel like I'm close to assimilating all the little facts that have been accumulating, but that I don't have my head wrapped around something crucial.

What I know:

1. ad8610 requires <= 26V power supply
2. opa627, opa637 require > 12V
3. 14.4V (12AAA) is within the calculator's range for the 24V elpac power supply
4. you (tangent) hint that the calculator can be fudged within a volt or two
5. 14AAA buys me 16.8V, which can be charged with a 26V PS for certain and maybe a 24V

How do I know/calculate what the fudge factor is between the battery board calculator and real world? Is there a way for me to calculate whether or not 14.4V is sufficient headroom for the burr brown chips, or, do I need to just build it and find out empirically? What makes a power supply good, should I decide to go shopping for a 26V PS?

And, I see that I do not understand the relationship between power supply voltage, opamp voltage requirement and headphone impedence -- what do I read to get my noggin wrapped around this concept?

 

[phaedrus]

And one more question:

Is the total draw of the PPA (12 buffers, biased into class A) ~100mA? Do I therefore expect to get 7-8 hours of off grid time with 800mA AAA batteries?

 

[Iron_Dreamer]

hey phaedrus, that SR325 sounds really incredible, let me know which opamp you prefer in your PPA for the 325, as I'm think about getting a PPA built some time in the future.

 

[phaedrus]

Quote:

Originally posted by phaedrus And, I see that I do not understand the relationship between power supply voltage, opamp voltage requirement and headphone impedence -- what do I read to get my noggin wrapped around this concept?

Is this article by tangent what I need to answer my own question?

 

[tangent]

Quote:

the calculator can be fudged within a volt or two

Yes. The slack is due to the fact that the dropout voltage of the regulator isn't a hard wall. First, the dropout voltage goes down as output current goes down, so you can raise R2 to get a bit more room between the voltage drop across the regulator and the regulator's dropout voltage. (See the regulator's datasheet for the dropout vs. output current graph.) Second, if you do run the regulator a bit into dropout, its current will drop. That graph I mentioned says the same thing. The only difference between these two modes of operation is whether you're limiting the current with R2 or relying on the regulator's design to limit the current. R2 is "cleaner", IMO, and it will give consistent performance as the battery pack voltage rises.

Quote:

How do I know/calculate what the fudge factor is between the battery board calculator and real world?

You can work it out mathematically with the datasheets for the batteries you're using and the regulator, plus details like the drops across the diodes and other things in the path. Or, if you have a bench supply, you can watch it happen: output current will start dropping as you lower the output voltage of the supply below the regulator's dropout voltage. I did this a lot when developing the battery board: I'd intentionally set the output voltage too low, then tweak it upward until the current didn't rise any more, then kept tweaking it upward as the battery charged. At the end of the charging cycle, you know the minimum charging voltage.

Quote:

should I decide to go shopping for a 26V PS?

You probably won't find one. Off the shelf supplies use multiples of 12 and 15V, for the most part. If you want a supply that's precisely 26V, you'll have to build one specially, or follow a 30V supply with a post-regulator to bring it down to 26V. This is the main reason we're planning on a custom power supply for the PPA: generic supplies don't have the adjustability one wants with the battery board. Don't hold your breath while waiting for it, though...

Quote:

Is the total draw of the PPA (12 buffers, biased into class A) ~100mA?

It can be.

Add up the quiescent draws of all the major components: the buffers, the op-amps, the LED, and the biasing current sources.

Quote:

Do I therefore expect to get 7-8 hours of off grid time with 800mA AAA batteries?

If the draw is 100mA, you would get 8 hours of run time from them if you drain them completely. If the amp starts clipping before the cells are drained, you won't get 8 hours of run time. Also, look at the cells' datasheet, if you can find it: as current rises, run time doesn't drop linearly across the whole scale. If you're outside your cells' linear range, you will get less run time than predicted by the mAh number.

Quote:

Is this article by tangent what I need to answer my own question?

It would help for you to read it, yes.