John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
Jump to Latest
#82,341
bcarso said:
Anyone remember the Burr-Brown modified TO-3 package buffer, the 3553? I see someone wants 200 US for one.
Click to expand...
I remember the LH0063 in a TO-3. The datasheet listed the little brother device LH0033 in a G12B metal can, and the LH0063 in a TO-3, as

Fast, and Damn Fast Buffers.

Sadly the datasheets that survived into the internet era, were scrubbed of this little piece of amusement.

edit- google searching for snoa725a takes you to app note 227 which discusses these (sans "Damn Fast")
 
Last edited:
#82,342
bcarso said:
Anyone remember the Burr-Brown modified TO-3 package buffer, the 3553? I see someone wants 200 US for one.
Click to expand...

I think I have one of these somewhere, several folks made them, power op-amps and power buffers for a period when folks paid $$$ for them. I posted pictures and schematics here a couple of years ago ironically most of the discrete chips that went into them are still around.
 
#82,346
bcarso said:
I wonder what the PJFETs in them were?
Click to expand...

I'm afraid I've probably tossed this stuff out. I don't recall there being PJFET's in the BB parts. The inputs were N diff-pairs IIRC (2N5xxx from Siliconix).

EDIT -Sorry I was thinking of the op-amps. Siliconix did have PFET singles in their catalog but I would have to find it to see if there were duals. Micropower and Siliconix regularly offered matched singles in waffle packs though.
 
Last edited:
#82,347
C'mon guys.

For line level stuff you need only TWO transistors and two diodes, or three transistors and no diodes plus an opamp to create a class A buffer with or without gain. A single resistor from the opamp output to the buffer output bootstraps the opamp into class A. Suitable SOT223 devices cost ~50c or so.

I published a circuit with measurements in low single digit ppm in 2009 on this. 10V pk into 200 Ohms at < 10ppm at 20 kHz, and ~2-3ppm into 600 Ohms. measured on an AP - sims said ppb.

(and no Waly, I am not 'promoting' myself here, just pointing out that there's no need to overcomplicate things)
 
Last edited:
#82,349
simon7000 said:
Attached is a plot of a string of twenty 2 ohm resistors fed via a 1,000 ohm unit. The frequency response is measured at each tap. The parts were on a solderless breadboard.

The second plot is a single 2 ohm resistor just clipped to the 1,000 ohm unit.
Click to expand...

Hi Ed
What do you want to show with these plots?
If I understand, the Y axis is the voltage drop across the resistors strip. It seems you are using a 3V source.
The increase in voltage drop at higher frequencies seems to be due to the inductance of the strip construction (*)
The difference btn the lowest cyan line (2 Ohm tap) of att.1 and the single line (single 2 Ohm resistor) of att.2 confirm this.

(*)it shows as a lot of inductance (~10 -15mH ?)
George
 
#82,352
gpapag said:
Hi Ed
What do you want to show with these plots?
If I understand, the Y axis is the voltage drop across the resistors strip. It seems you are using a 3V source.
The increase in voltage drop at higher frequencies seems to be due to the inductance of the strip construction (*)
The difference btn the lowest cyan line (2 Ohm tap) of att.1 and the single line (single 2 Ohm resistor) of att.2 confirm this.

(*)it shows as a lot of inductance (~10 -15mH ?)
George
Click to expand...

George,

What I think I am showing is how bad the solderless breadboard actually is! If it can ruin test results with an impedance of 2 ohms, it has to be horrible. I read the increase level at HF as leakage capacitance.

Using as few and hanging in the air connections I got the much better results as shown. I did drop the input voltage from 5 to 1.25 volts peak.

Attached is the final results of a three way loudspeaker with base line graphs of 4, 8, 12, 16,... resistors without the breadboard.
 

Attachments

  • PD7641-215.jpg
    PD7641-215.jpg
    633.4 KB · Views: 205
#82,353
If you look it says measured on an Audio Precision...
Click to expand...
for that matter clever technique can wring a lot out of better < k$ soundcards like the ESI Juli@

it is more tedious to get good results compared to dedicated equipment like AP - but its also unlikely to err on the "too good to be true" side - more likely to hit equipment limitation with a soundcard
 
#82,354
We had two AP's in the Tokyo lab - I used the SYS2722 (I think that's the model name) to do the measurements on the buffer.

http://hifisonix.com/wordpress/wp-c...niversal-Small-Signal-Class-A-Buffer-V1.0.pdf

Interestingly, there was a big clear-out of the lab and a lot of gear that was fully depreciated was scrapped off. I got a great 200 MHz Philips analog scope (which was completely destroyed last year by the shippers en-route from Taiwan to the UK - another interesting story).

Well, back to the Tokyo story: I ran like hell up the stairs to the lab, my heart set on the old AP which still worked perfectly. Some other guy got there first unfortunately. Philips was a big player in CD's, Amps (Marantz, Automotive, Consumer class D etc) so the place was well set up. Stax headphones etc

I'm using a Focusrite Scarlett Solo now - with some preconditioning gain its good for about 6 or 7 ppm.
 
Last edited:
#82,358
Bonsai said:
Well, back to the Tokyo story: I ran like hell up the stairs to the lab, my heart set on the old AP which still worked perfectly. Some other guy got there first unfortunately. Philips was a big player in CD's, Amps (Marantz, Automotive, Consumer class D etc) so the place was well set up. Stax headphones etc
Click to expand...

The moral I'm taking away from this is keep up my exercise regime. 😀

Ugh on the scope, nor am I surprised by the buffer's performance. The LM4562 is stupid good and a tidy little bit of discrete buffer oomph isn't going sully that too much.
 
Status
Not open for further replies.
Top Bottom