[primer]

Quote:

Originally Posted by skyskraper i put my pimeta's in to the standard serpac h65.

skyskraper, did you get the serpac case in oz?

 

[skyskraper]

you can get em from jaycar for $8.95 linky linky

 

[primer]

Quote:

Originally Posted by skyskraper you can get em from jaycar for $8.95 linky linky

Thanks skyskraper, I did see the one from jaycar but wasn't sure of the build and quality as compared to the proper serpac.

 

[skyskraper]

theyre really good, defn reccomended and nice and cheap too!

 

[mono]

Quote:

Originally Posted by rcgrant Shouldn't a PIMETA have comparable battery life to a MINT, seeing as how their circuits are virtually identical?

It could be similar, except the Mint lacks the ground channel, so add the BUF634 [in low bandwidth is (1.5mA ?) ] and *whichever* GND opamp you chose too... usually that's about 8mA more at a minimum.

8mA+ is a pretty high percentage of the total (all other things being equal), or to look at it another way, Mint has 2 channels and Pimeta 3, with each channel having similar potential current draw so Pimeta uses almost 50% more.

HOwever, some of us (ok, maybe just me) see the Mint as a lower-cost option than the Pimeta, and might use lower-cost (& likely lower-current) opamps. Pimeta can use a browndog for a pair of single-channel opamps on L & R channels- even more current. I suppose it's technically possible on a Mint too, but if you start getting elaborate with one then suddenly you need a bigger chassis and it's size advantage isn't there anymore.


A Pimeta builder might be inclined to use higher frequency opamps, more current. Sure, nobody HAS to build it like that, but many of 'em are easily > $100 when finished so one might be aiming for least compromises possible in that price-range because it's about as big as it gets before it can't be considered portable anymore (sorry PPA fans, I don't consider PPA portable any moreso than I could pick up my home stereo receiver and carry that somewhere too, ignoring the power source). That is, if they don't need it to run all day.

The opamp & buffer spec sheets will list current consumption. You can estimate this with good enough accuracy to plan the design providing you keep the LED current minimized too.

 

[tangent]

Quote:

Most people using 18V are simply wasting space and a battery.

That statement is true and false, depending on the op-amp. The OPA627 I wouldn't run on less than 10V, for instance. If you're using the AD8610/20, then the benefit of 18V is certainly questionable.

Quote:

Is the current draw so great that a 9V will only last for a few hours?

It depends on how you configure the amp.

There really is no rocket science here. As long as you supply the op-amp enough voltage to avoid clipping before the battery is dead (which is the case with the AD8610/20 and a 9V rechargeable and typical headphones), you simply add up the quiescent currents and divide that into the mAh rating of your battery.

Since you know your run-time goal, you can go at it the other way. Let's say you're using a 170mAh Plainview, and you want to get 8 hours out of it. We'll be a bit conservative and bump that up to 10 hours, which also makes the numbers easier: we can tolerate a 17mA quiescent drain. The max quiescent rating on the AD8610/20 is 3mA per channel, so there's 9mA gone right there. Three BUF634s with no bandwidth adjustment are another 4.5mA. The LED is another 1mA or so; call it 1.5mA to round things out. That leaves us with 2mA left to play with, so let's throw it towards class A biasing, 1mA per channel. Done.

That wasn't so painful, was it?

Quote:

GND opamp you chose too... usually that's about 8mA more at a minimum.

Ah, very few op-amps are so hungry as that. Most that you'd consider for portable use in the first place are in the 3 to 6mA range.

 

[mono]

Tangent, have you tested the OPA627 with lower voltage when it's behind a buffer? I built one of my Pimetas with OPA627 and (IIRC, I will recheck this) it ran ok down to around 5V, rail-to-rail. The last couple of volts drop might've been questionable but it certainly sounded good at +-4V (battery pack @ 8V & continuing to drop).

Later I pulled the OPA627 pair and put OPA2107 in L & R of same amp which is never above 8.7V with it's current battery and it too sounds good. I don't recall if I've let the battery run down as far and listened to OPA2107 but I expect similar results as with the OPA627s.

It could be that I simply can't discern the slight audible difference at around 10V, but there was a definite cliff of distortion much lower, about half that voltage. I have ran same amp with 12V NiMH pack and 24V bench linear (fairly similar to your STEPS) rail-rail for comparision. I think the buffer(s) greatly reduce the minimal voltage allowance.

However, I should add that it wasn't quite a stock Pimeta, it has quite a few Tantalum and Ceramics for decoupling stuffed all over the PCB. I'll have to take a picture of it, quite an ugly monster it is after I got done experimenting with it.

By the 8mA figure I meant that to be both the opamp and buffer, a rough total.

 

[mono]

I"m now wondering if we can simply take the rating of the battery at it's face value or might have to de-rate it some.

A pseudo-9V NiMH is probably rated till it drops down to 1.0V/cell, if the same methodology as single 1.2V cells is used. If the Plainview is the one with the 7 cells in it, that provides a little more margin but still, to get the 170 mAh it'd be down to 7V, which could be problematic if there is a distortion introduced in some opamps at certain voltage thresholds? AD8620 would handle it best based on your testing in the CMOY?

I'll give my Pimeta another listen at lower voltages but I'm pretty sure I was critical when I judged it ok @ 8V, before I ever chose a case and battery pack for it.

 

[tangent]

Quote:

Originally Posted by mono have you tested the OPA627 with lower voltage when it's behind a buffer?

No, that comment comes mostly from snobbery. I figure that a chip of that caliber should be treated lavishly, so that it will feel inclined to return the favor.

Quote:

A pseudo-9V NiMH is probably rated till it drops down to 1.0V/cell

0.8 or 0.9V/cell is a more common cut-off point in my reading. Think about it: why wouldn't the marketing department choose the lowest practical voltage they could? And if you think the datasheets aren't written by the marketing department, you're fooling yourself.

Quote:

If the Plainview is the one with the 7 cells in it

No, it's a 9.6V unit, hence 8 cells.

Quote:

to get the 170 mAh it'd be down to 7V

You're essentially correct: it all comes down to the clipping question.

As I said above, in my experience, you can fairly be said to have fully discharged a Plainview with the AD8610/20 and common headphones before clipping sets in. In fact, you're so near the end of the battery's life, that it doesn't so much go into clipping as die outright: the terminal voltage is falling too fast for you to hear much clipping at all before it's gone.

 

[mono]

Quote:

Originally Posted by tangent No, that comment comes mostly from snobbery. I figure that a chip of that caliber should be treated lavishly, so that it will feel inclined to return the favor.

I might've opted for a little more voltage but at the time I could only squeeze 6 prismatic cells into a Serpac H65. To my ears it sounds same as operation at 24V and the prismatics gave me a 900mAh pack, so at least for me it was a good choice.

 

[mono]

Frankenstein?

 

[Heady]

mono, how do you charge those cells?

 

[rcgrant]

Thanks for the calculations, tangent. That configuration sounds perfect for my purposes.

How do you have the gain set on your PIMETA?

 

[tangent]

Quote:

Originally Posted by rcgrant How do you have the gain set on your PIMETA?

With resistors. How did you plan on doing it?

g=2

 

[mono]

Quote:

Originally Posted by Heady mono, how do you charge those cells?

No different than anybody else? With the power jack on the side, it's a std. 5.5/2.5 mm. If you meant the supply, I can plug just about any unregulated wall-wart into it for charging only but I have a couple warts now with regulated max voltage and another with low current that can trickle charge it. When that pack wears out (in a few years) I'll put an inline 9V snap-on type connector on it, I just didn't have any spares at the time.