[spongezone]

I'm building a MINT amp right now, and I'm trying to figure out how much bandwidth the buffer should be set to used with an OPA2277. Would wide open at 180mhz be too much? Also, what is the formula to figure out what resistor to use to select the bandwidth? I checked the Quiescent Current vs. Bandwidth resistance chart in the datasheet, but I can't make much sense of it.

 

[__redruM]

Tangent's mint site should give you a good starting point. If you're not concerned with longer battery life, use 220ohm per buffer on the bandwidth resistor. You can go lower, but it doesn't seem to help.

 

[individual6891]

Quote:

Originally Posted by spongezone I checked the Quiescent Current vs. Bandwidth resistance chart in the datasheet, but I can't make much sense of it.

The lower the resistance, the higher the current draw

 

[n_maher]

I'd say throw a 1k resistor in for R11 and see if you like the way the amp sounds. It should provide a nice balance between current draw and buffer performance. The MINT was never intended as a super high end amp, don't get me wrong, I use one every day and it sounds good. But it get's blown away by the likes of a maxed PIMETA. Of course that same PIMETA that I'm using for comparison only had a 4hr run time on batteries. My MINT generally goes for 20 or so.

From my experience going from no resistor (R11 open) to 1k made an audible difference in the sound of my MINT. Additional gain (slight) was made moving to 220ohm but at a further reduced battery life. I left it in there because I was too lazy to go back to 1k.

Finally, I would strongly suggest that you do not jumper R11 as this will draw power like crazy for no purpose. It's in Tangent's MINT docs (under Tweaks) that people have found that this actually sounds worse than having say a 220ohm resistor in there.

 

[amb]

The OPA2277 has very poor bandwidth (GBP of only 1MHz and slew rate of 0.5V/uS). This is just as bad as the old 741 opamp and I consider it to be unacceptable for serious audio use. At any rate, given this you will not gain anything by operating the BUF634 in wideband mode. With a better opamp the situation would be quite different.

 

[KarlDL]

My first HP amp is a hybrid of MINT, PPA, and M3 techniques. Right now, the BUF634s are fed by OPA2132s. I tried NE5532s but the input offset current was a problem. I'm using 100 Ohm "bandwidth" resistors, chosen not for BW but rather for quiescent bias current. Can't comment on whether 220 or 1K would make a sonic difference. This just seemed to me to be a reasonable starting point for a wall-wart-powered amp (32VDC supply from 24VAC wart via LM317).

b/t/w, the dual op-amps don't sound quite as good as the OPA627s in my original breadboarded version. But this is not intended to be an amp of ultimate sonics, rather, a test-bed for some general principles.

 

[tangent]

spongezone, just checking: do you mean the OPA2277, or in fact the OPA2227?

 

[spongezone]

I had a few OPA2227s left over from another project. I guess I should get a 2277 instead.

KarlDL, I'm using a dual opamp due to space and power limitations. I'm putting the amp on half of a RS 276-0150 protoboard.

 

[bg4533]

Quote:

Originally Posted by spongezone I had a few OPA2227s left over from another project. I guess I should get a 2277 instead.

Did you not read what was posted or did you confuse the numbers?

The OPA2227 is the better opamp.

 

[amb]

Quote:

Originally Posted by bg4533 The OPA2227 is the better opamp.

The OPA2227 is better, but only a little. The slew rate of 2.3V/uS is still nothing to write home about. The (also inexpensive) OPA2134 a much better substitute.

 

[spongezone]

Quote:

Originally Posted by bg4533 Did you not read what was posted or did you confuse the numbers? The OPA2227 is the better opamp.

Oh, oops. Mistyped in my last post.

I had a few 2277s left over. I'll get a 2227 or 2134.

 

[mono]

IMO, one builds a Mint for the smallest, most portable design without drastic compromises (like no buffers, a la CMOY). Along with that idea is that towards such a goal the battery would be small, relatively low capacity. For these reasons the low-bandwidth mode is the only that makes sense (to me).

On the other hand, I would choose better than OPA2277 regardless of the power difference, as it may relatively make more of a difference. Since this is a Mint and surface-mount, I think the safest route is *too much* amp rather than too little... unless of course you'd just move on to a better portable amp as next project and thus might not end up using the Mint much. Personally, I'd probably choose OPA2107. Although it has higher voltge *requirement*, the buffers relax that and it should work from 9V.

 

[spongezone]

Quote:

Originally Posted by mono IMO, one builds a Mint for the smallest, most portable design without drastic compromises (like no buffers, a la CMOY). Along with that idea is that towards such a goal the battery would be small, relatively low capacity. For these reasons the low-bandwidth mode is the only that makes sense (to me). On the other hand, I would choose better than OPA2277 regardless of the power difference, as it may relatively make more of a difference. Since this is a Mint and surface-mount, I think the safest route is *too much* amp rather than too little... unless of course you'd just move on to a better portable amp as next project and thus might not end up using the Mint much. Personally, I'd probably choose OPA2107. Although it has higher voltge *requirement*, the buffers relax that and it should work from 9V.

Hmm, how much battery life could I expect from a Mint with an OPA2107 and 2 BUF634s. My ultimate goal is to fit the thing along with 1 9v in one of those Altoids Gum containers, but if itll only last a few hours, ill go with 2 9v's in a regular Altoids container. I might just go with the 2 9v's in series and use a higher voltage Op Amp.

 

[mono]

I'll ignore the total power consumption of the rest of the amp, including buffers since they may be a constant regardless of opamp chosen.

OPA2227 has 3.8mA/channel. OPA2107 is 4.5mA. The difference is slight, alone not enough to warrant moving to larger case and doubling batteries. However, as with any amp I suggest you try it out before putting it into a case. For high volume on hard cans you might want the higher voltage? The opamp you choose may not matter in this regard since it's buffered. I didn't find it necessary on a Pimeta but YMMV.

OPA2277 though... I'd question whether the compromise was worthwhile, I'd almost as soon use a pair of better opamps without buffers if battery life was THAT important, though that's not a mint anymore!

 

[spongezone]

Quote:

Originally Posted by mono I'll ignore the total power consumption of the rest of the amp, including buffers since they may be a constant regardless of opamp chosen. OPA2227 has 3.8mA/channel. OPA2107 is 4.5mA. The difference is slight, alone not enough to warrant moving to larger case and doubling batteries. However, as with any amp I suggest you try it out before putting it into a case. For high volume on hard cans you might want the higher voltage? The opamp you choose may not matter in this regard since it's buffered. I didn't find it necessary on a Pimeta but YMMV. OPA2277 though... I'd question whether the compromise was worthwhile, I'd almost as soon use a pair of better opamps without buffers if battery life was THAT important, though that's not a mint anymore!

Alright, thanks for the info. I'll go with a 2107.