[cire]

less than 10mv with volume on full tilt. read this: http://www.mb3k.com/hosted/mmmpint.pdf

there's a table with gain from 2 to 11 saying which resistors to use and where.

 

[threepointone]

oops, i left out the "how" before; i know what resistor values were recommended, i was just wondering how those values were chosen

 

[Teerawit]

Quote:

Originally Posted by threepointone oops, i left out the "how" before; i know what resistor values were recommended, i was just wondering how those values were chosen

To yield an equivalent resistance (of R3 and R4, which are in parallel) of 1k to balance the input and output impedances of the opamp so it doesn't oscillate and blow up never to be heard of again.

 

[BavariaBarbarian]

Quote:

Originally Posted by Filburt and I've done enough experimentation already that my wallet is starting to put up a fight.

I can relate to that one.

Quote:

Originally Posted by Filburt Tangent and amb probably have more experience with this than I do, so they'd be good folks to ask.

I did ask Tangent in another Thread and haven't got a reply.
I might shoot him or Amb a PM, but I guess I'll just try for myself when I finally get around to building the pint.
This is DIY after all.

AFAIK you only can fit >4 AAAs and the pint into a mint if you use solder-tabbed ones and build a pack.
But I'd love to be wrong on this...

 

[Filburt]

Er, I'm not sure how it'd only be 4 AAAs, but maybe it is under your method. I just tried to guesstimate it by holding a AA battery in the space open for batteries and it looked like I could fit 4 AAs, so I figured at least 6-8 AAAs would fit.

 

[tangent]

Quote:

Originally Posted by BavariaBarbarian If I did the maths correctly 4V will give me at least 1.1V voltage swing

I'm not sure how you can get that number, based on the information in the AD8397 datasheet.

What you can find is that with a 5 V supply, you can put about 4.75 Vp-p into a 25 ohm load, which is quite impressive. That's just under 1.7 Vrms, which is plenty for most headphones.

Quote:

Could you please elaborate on the difference in sound? Have you actually gone as low as 4-5V?

I'm having trouble with my 4xAAA configuration at the moment, and haven't been back to revisit it. Too many other things going on right now, unfortunately.

 

[amb]

I have not done any serious listening at low supply voltages, so don't have any comparison data. But, one thing I did notice with the AD8397, is a slight increase in quiescent current in the neighborhood of 5-6V. Normally at 18V+ the whole circuit draws about fourty-something mAs (maybe close to 50mA), but at the low voltages it increases to over 60mA.

So, something to think about if you're trying to conserve battery life.

 

[BavariaBarbarian]

Thanks for the input Tangent and Amb!

Quote:

Originally Posted by tangent I'm not sure how you can get that number, based on the information in the AD8397 datasheet.

Ah well... Calculator aided guesswork.

I guess the trouble with the 4AAA version is regarding stability?

From the datasheet I wouldn't have expected the increased current draw at lower voltages, but the reverse. Of course that'll offset the higher capacity in part, but ~800mAh AAAs still would last 10+ hours.

I guess I'll just try it. A two AAA holder is about the same size as a 9V Block, so it's easy enough to switch to 2x9V if it won't work.
Unfortunately this'll have to wait until summer break.

 

[fierce_freak]

Quote:

Originally Posted by blueworm See http://headwize.com/ubb/showpage.php?fnum=3&tid=6096 Then for a complete explanation see:- post #81 http://www4.head-fi.org/forums/showt...=173192&page=5

I've just recently gotten back into doing electronics work...there's a pdf floating around saying to do a cut and make a jump. i understand that well enough, but in the schematic it seems that the cut is labeled jump and the jump is labeled cut. Am I looking at it incorrectly?

 

[MisterX]

Quote:

Originally Posted by fierce_freak Am I looking at it incorrectly?

Sounds like it.
Maybe this will help

 

[dhp]

Quote:

Originally Posted by MisterX Sounds like it. Maybe this will help

doesn't help him, I'm sure. I think you're reading it wrong. Red is the tracing, and black is the nothing

 

[fierce_freak]

thanks MisterX and DieInAFire...that's what I was doing, reading black as traces. I didn't look at it closely enough or I would have realized it, haha.

 

[threepointone]

Quote:

Originally Posted by Teerawit To yield an equivalent resistance (of R3 and R4, which are in parallel) of 1k to balance the input and output impedances of the opamp so it doesn't oscillate and blow up never to be heard of again.

i know, but i was just wondering where this magic number 1k came from. a current is forced through the resistors in the noninverting input from signal to ground, so (at full volume) shouldn't the resistance to ground just the 100k resistor and 1k + 10k pot in parallel = ~10k in order to balance the noninverting and inverting inputs? So that would also mean that the DC offset can vary as you change the volume or with different source output impedances.

I'm almost certain there's something wrong with my thinking, but i can't quite figure out what it is right now

 

[phobus]

Just built my first PINT, standard configuration with 2 8.4v GP rechargeable batteries. I had enough room for a 1000uF Panasonic FM cap. Sounds great with the AKG 501s. Beats the pants off my portable Pimeta (8620) and MINT (8066).

Is there supposed to be a temperature difference between the "regular" PINT and the minified version?

 

[tangent]

Quote:

Originally Posted by threepointone i was just wondering where this magic number 1k came from.

It's just picked at semi-random by the Mini3 team. It's as good a value as any, and it's a value that hobbyists that stock parts is likely to have already.

Quote:

a current is forced through the resistors in the noninverting input from signal to ground, so (at full volume) shouldn't the resistance to ground just the 100k resistor and 1k + 10k pot in parallel = ~10k in order to balance the noninverting and inverting inputs?

Your mistake is in treating the pot as a fixed resistance. From the noninverting input's perspective, the pot's resistance varies. At minimum volume, the noninverting input only sees the 1K resistor between it and ground.

This variability between 1K and 11K is why I don't recommend this configuration as a default: there is only one point where you truly balance the input bias currents, so it's possible that DC offset rises too high at some volume levels. At the same time, though, I don't speak out and denounce it because it can work out okay. It just requires more care in parts selection and testing.