LHoo33 and 63 "Damn Fast Buffers"

(Short version: I'm wondering how to prevent from accidentally frying these devices, as they are hard to come by...they have no output protection.)

Longer version:

Ive been very happy with the circuit below used as a headphone amp and/or hi-fi preamp...this using LHoo33's ( with OPA602 or other opamps).
(btw, the circuit sounded better, with the "Rsc" output resistor...outside the loop, not inside like my schemo/datasheet schemos.)
pic:
http://i5.photobucket.com/albums/y177/Midiot/nonparallelLH0033.jpg

I just recently bought a pair of LHoo63's ($$), and I'm concerned about the devices' lack of short circuit protection.
Does the output resistor in my circuit above act as protection ?
Or should I employ the technique below, as in the datasheet(s) ?
..or is there a better way ?
I want to preserve as much current on the output, to drive my 32 ohm headphones.

From the datasheets:

"" Because of their high current drive capability, the LH0002,
LHoo33 and LHoo63 buffer amplifiers are suitable for driving
terminated or unterminated coaxial cables, and high current
or reactive loads. Current limiting resistors should be
used to protect the device from excessive peak load currents
or accidental short circuit. There is no current limiting
built into the devices other than that imposed by the limited
beta of the output transistors. ""




http://www.national.com/an/AN/AN-227.pdf#page=1
http://rocky.digikey.com/WebLib/National/Web%20Data/LH0033,%20LH0063.pdf


=FB=

Actually the answer is in the datasheet you linked (fig. 22)? 47 ohm in series with the output will limit the current to about 300 mA with 15 V supplies. That's low enough to prevent failure with accidental shorts of a few minutes but will still overheat the driver if you plan to have it shorted for hours with full signal applied. Depends on how you use it--if that's just your personal amp for CD listening at home it will suffice, but if you intend it for studio use where other people may use it as well it might be wise to include the full protection circuit--preferably that shown in fig. 21. In any case you should replace the 47 ohm resistor with a 5 W part, or it will act as a thermal fuse before the driver burns.

BTW, the circuit sounded better, with the "Rsc" output resistor... outside the loop, not inside like my schemo/datasheet schemos.

Click to expand...

I'd be careful with that conclusion. Having the resistor outside the loop will cause varying frequency response with different headphones. It may sound good with one but hurt another one.

Samuel

I have some of these "fast" and "damn fast" Ic's I bought in the 80's.
I'll keep and eye on them in case you cook these!=)
I get misty-eyed when I pick up my National Analog Databook....sniff..

[quote author="Samuel Groner"]Actually the answer is in the datasheet you linked (fig. 22)? 47 ohm in series with the output will limit the current to about 300 mA with 15 V supplies. [/quote]
[quote author="Freq (me)"]Does the output resistor in my circuit above act as protection ?[/quote]

I thought this might be the case, thanks.

[quote author="Samuel Groner"]

BTW, the circuit sounded better, with the "Rsc" output resistor... outside the loop, not inside like my schemo/datasheet schemos.

Click to expand...

I'd be careful with that conclusion. Having the resistor outside the loop will cause varying frequency response with different headphones. It may sound good with one but hurt another one.
Samuel[/quote]

ooh,,, point taken.

[quote author="Larrchild"]I get misty-eyed when I pick up my National Analog Databook....sniff....[/quote]
I got misty-eyed this weekend, having to pay $60 for each LH0063....
...but this was my only source, and, I'm making this nice preamp for myself, so....
Actually I also have some of the LHoo32 opamps shown in the datasheet schemo above. I don't know whether these would sound different than a more modern opamp ???? Someone mentioned that an opamp's sonic character is less discernible when used inside a buffered loop. This is, of course, one thinks they can hear such a thing.
For me, I can't taste whether an apple was picked from the east...or west side of the tree.
I liked the OPA602's because of the low noise, and are easily trimmed to essentially 0v offset.
But then again, nostalgia might nudge me to try the LHoo32's. I suppose I could trim them as well.

=FB=

[quote author="Samuel Groner"]In any case you should replace the 47 ohm resistor with a 5 W part, or it will act as a thermal fuse before the driver burns.

Samuel[/quote]

So in the case of a 5w resistor, potential "sound quality" would be preserved by using what type resistor ?

I have:
High quality wire-wounds (Shallco, Dale, Sprague "Blue Jackets").
Non-inductive wire-wounds (Sprague "Koolohm", Dale NS-5, Kelvin KMN500).
Carbon Film, Metal Oxide power types



=FB=

As it stands now, the output will be shorted briefly 1-2 seconds. I am employing a muting circuit that shorts output to ground during power On/Off.

http://sound.westhost.com/project104.htm

=FB=

EDIT: or these?
http://www.caddock.com/Online_catalog/Mrktg_Lit/MP9000_Series.pdf

EDIT 2: silly me...
I actually have several of these @ 125 ohm, (8watts), and paralleled, gives me 62.5 ohm (target is 64 ohm)
http://www.vishay.com/docs/63012/vpr22021.pdf

Increasing the resistance of the limiting/protection resistors in series with power supply rails can provide protection against longer duration shorts, while capacitors there can supply short term current for audio peaks. This will have less sonic impact than any series resistance in output.

While perhaps a little lowering the water in lieu of raising the bridge, what about using the relay to disconnect output from the driver? While you will have a relay contact in series with the audio path this should be OK with a decent relay.

Note: I would never design a commercial product without some series resistance in the output but for driving headphones you want to keep it small relative to headphone impedance.

JR

[quote author="Larrchild"]I have some of these "fast" and "damn fast" Ic's I bought in the 80's.
I'll keep and eye on them in case you cook these!=)
I get misty-eyed when I pick up my National Analog Databook....sniff..[/quote]

I remember the polish opamp from that time period.