[ozshadow]

n/m

 

[Teerawit]

Quote:

Originally Posted by ozshadow Jameco has BUF634P in stock which should work if you bend the lead tips flat and level.

Too big and mechanically unsound.

 

[quicksilver96]

Quote:

Originally Posted by ozshadow Jameco has BUF634P in stock which should work if you bend the lead tips flat and level.

Not gonna happen, don't do this.

Here is a size comparison for anyone unfamiliar with the size of a DIP chip vs SOIC pads and available board space:
Image Gone

 

[motherone]

They do have DIP to SOIC adapters, but you're going to be paying as much as you are for the buffers for them, IIRC.

 

[ATAT]

More like.. you'll be paying $20 for each

rsabo (i think) made some intersil DIP8 -> SOIC adaptors but I havent made em yet.

 

[ozshadow]

If you are very tight in your assembly, I *think* you can slide all this into a C1201 hammond case. Batteries need to go end to end on one side and the board needs to be inserted upside down, but it should fit, and definitely would with a light bit of sanding here and there. Should leave a tad bit of room for a trickle charger.

 

[ozshadow]

Tightly packed C1201 case to scale. There is no room for sloppiness here.

Similar to this: http://www6.head-fi.org/forums/showp...&postcount=164

I do not have this setup to play with, but I am guessing the board could be angled a bit for slightly more space

 

[ozshadow]

Also, for those of you with a built version, is there enough room to use 6x11 caps on the opamp rails instead of the 5x11 listed ? This must be absorbed in width of the board and not pushing the other caps outward.

The reason for this is that I need 16v caps instead of 10v for the charging/power circuit which will be splitting approx. 27 volts.

There is a Panasonic FM cap in 120uf 16v that is 6x11 - one extra mill in diameter.

 

[pddjsteve]

Quote:

Originally Posted by ozshadow Also, for those of you with a built version, is there enough room to use 6x11 caps on the opamp rails instead of the 5x11 listed ? This must be absorbed in width of the board and not pushing the other caps outward. The reason for this is that I need 16v caps instead of 10v for the charging/power circuit which will be splitting approx. 27 volts. There is a Panasonic FM cap in 120uf 16v that is 6x11 - one extra mill in diameter.

I'm out of town and don't have the board with me, but from memory it was a pretty tight fit just with the 5mm. You could probably mount the caps a little off-board and bend them out of the way, though.

 

[quicksilver96]

Any takers on using the LMH6321 in this project yet?

I was looking at the data sheets for this buffer and had a question about the GND pin (5). Does anyone think this needs to be connected to the ground plane if the current limiter and error flag pins are unused? Looking at the simplified schematic (fig 1) I would say that it doesn't. In the subsection "Single Supply Operation", the guide mentions only one way to attach the ground pin and does not specify conditions when this connection is necessary. Also, I see there is no way to implement the current limit without some degree of hackery and the trace to pin 2 should be cut, or the pin should be removed from the chip, as this is a feedback loop in the PPAS circuit. In unlimited configuration the chip (open CL pin) will be able to provide up to 700(!) mA of current, though this should be avoided and it would seem that the CL pin provided any means of short circuit protection. Lastly, mechanically disabling the EF is unnecessary but there should not be anything bridging the high bandwidth jumper. Correct?

 

[ATAT]

Built my PPAS today, LM6171, LM6172, cant give you good impressions yet since ive had a total of about 10 minutes listening time, but BUF634/LM6171/LM6172 is deemed stable.

LM6321 had issues yesterday, tried em first and they started dumping a ridiculous amount of current onto the output didn't realize the output jack was shorted via a thin wire

So I'll have to try again.

Quicksilver - the GND pin does not need to be connected if current limiter is not connected. the buffer will operate fine as long as you have a split supply..

As i said in post #1, pin #2 should not be soldered on for this to work..

Lastly, there's no reason why I would implement current limiting.. of course there's the off chance that something like the aboe (a small wire short) may happen, but its not destructive to the amp in any way.. current limiting is pretty bad for many reasons and I'd rather not have it at all, which is why provisions for current limiting were not included.

 

[dhp]

sounds good? try grabbing some l3ks in japan and you'll be set

 

[quicksilver96]

Cool. I didn't think that current limiting was necessary given the load, just that it might deliver some type of short protection. Good to know about the ground pin, there was really nothing in the docs saying that it needed to be connected, and the only time it showed up on the schematic was for an internal opamp in the CL circuitry.

Were you just running the BoM config or did you make some bandwidth adjustments (R5, R6) with that fast chip? BUF634 in wideband?

 

[ATAT]

Wideband mode is undocumented - jumper to the left of the right buffer and below the pot is the bandwith jumper.. jumpering that gives wide bandwith mode while keeping it open leaves it in low draw mode

quicksilver - I havent got a LMH6321 config working yet so my post was speculation from the datasheet. the simplified schem shows ground being used only as a reference for the error pins (makes sense, since that error signal's gotta be referenced to something) and the rest looks just like a beefed up BUF634 (and therefore not needing a reference point midway the supplies..

For the LM6171 and such, no adjustments were made I will put the appropriate resistors to match bandwiths but the amp is stable at stock configuration.

 

[rsabo]

Quote:

Originally Posted by ATAT Wideband mode is undocumented - jumper to the left of the right buffer and below the pot is the bandwith jumper.. jumpering that gives wide bandwith mode while keeping it open leaves it in low draw mode quicksilver - I havent got a LMH6321 config working yet so my post was speculation from the datasheet. the simplified schem shows ground being used only as a reference for the error pins (makes sense, since that error signal's gotta be referenced to something) and the rest looks just like a beefed up BUF634 (and therefore not needing a reference point midway the supplies.. For the LM6171 and such, no adjustments were made I will put the appropriate resistors to match bandwiths but the amp is stable at stock configuration.

I might be missing something, but is there a similar jumper to the left & ground buffers? (I'm going by the pictures on the website, I've been away so I don't have my boards yet)