DIY Curve Tracer for PC

Status
Not open for further replies.
Jump to Latest
#422
Thinking of purchasing one of these

guys I have never used a curve tracer but am a quick study! I have BP ouputs i want to match and some smaller driver and signal transistors....this should do the job...

the question is did anyone adapt to measure/match op amps? this would be very helpful for what i need to do..


Thanks

Lawrence
 
#423
What are you trying to match on opamps? While there are curve tracer methods for opamps you can do the same with a few simple test fixtures. A lot depends on what you want to know and what data sheet factor you want to check. Many datasheets, especially older ones include test circuits and that's a good starting point. You would need a good dual adjustable bench supply, a suitable signal source and whatever is appropriate for measuring the target parameter, a meter, oscilloscope or possibly distortion analyzer.
 
#425
Probably the easiest and best would be the method on page 7 of this app note: http://www.ti.com/lit/ds/symlink/opa2604.pdf . What it does is reduce the available gain to drive the feedback network so the intrinsic distortion is reduced less by the feedback.

There are variations in the datasheets for other low distortion opamps. The distortion measured is directly proportional to the linearity.

Usually this turns out to be useful only for finding defective parts. Good IC's are really uniform making the whole effort tedious and somewhat pointless.
 
#426
Thanks

Hi Thanks but I was hoping that i could use the curve tracer to test a bunch of these opamps...

to be honest I do not trust opamps and want to match gain and linearity/distortion....please remember this is for an audio device 2channel etc..



Lawrence
 
#427
Lawrence,

You set the gain of a opamp by determining the feedback ratio. Accuracy is a function of the resistor ratio accuracy Usually you use 1% R's for this. For very precise ckts you use even higher precision R's, but they are not generally needed for audio work because 1% gain accuracy is more than enough for most audio ckts.

BTW, what do you not trust?
to be honest I do not trust opamps
Click to expand...
Gain set by you R selection.
Linearity/distortion. You do not trust what TI specifies, good luck proving them wrong. I do not think that you can afford the equipment necessary to make a claim like that.

Rick
 
#429
Its true that if you need either really precise gain (.0001% or better) or very large gain more that 120 dB the open loop gain will be an issue. It terms of nonlinearity the opamps of a given type will be very similar. These are usually controlled by passive R's and C's on the chip.

The manufacturing process is designed and optimized to yield really consistent parts. Most IC processes are more controlled than discrete parts. Any significant variations would make a part with 1 billion transistors impossible to make and Intel makes lots of them.
 
#434
Finally got to use mine

But not without an issue appearing. my N MOSFET curve looks fine but my P MOSFET curve has issues. Any suggestions? Possibly a faulty relay?
 

Attachments

  • MOSFET Curve.png
    MOSFET Curve.png
    40.5 KB · Views: 813
#435
Yup. something hairy going on, looks to be noise, could be bad connections.
How do pnp bjt's measure?
Lawrence, I should apologize for this statement
Linearity/distortion. You do not trust what TI specifies, good luck proving them wrong. I do not think that you can afford the equipment necessary to make a claim like that.
Click to expand...
After reading Bob Peace's article's, it would be interesting to test for opamp linearity, as he has done. Bob Pease was a magnificent contributor to electronics.
As in traditional TI fashion, they removed the authors name from their updated app note, a shame really.
 
Last edited:
#436
rsavas said:
Yup. something hairy going on, looks to be noise, could be bad connections.
How do pnp bjt's measure?
Lawrence, I should apologize for this statement
After reading Bob Peace's article's, it would be interesting to test for opamp linearity, as he has done. Bob Pease was a magnificent contributor to electronics.
As in traditional TI fashion, they removed the authors name from their updated app note, a shame really.
Click to expand...

its alright don't worry about it...I think its important to test all components in a piece of sensitive electronics including op amps etc..

Lawrence
 
#438
#439
Curvetracer failed

Hi all:
I tried emailing locky Z yesterday but no reply yet. I have a curvetracer 3 months old...and the PC s'ware config 'device not detected' no longer detects it although the PL2303 port is detected and the curvetracer relays click once the USB is connected. I am using Com4 as the virtual port as indicated by the device manager, Win XP SP3. No yellow flagged driver probs.

Reinstalled the supplied PL2303 driver (2013) with no probs.

Any thoughts on what to do?😕

thx
Ancel
 
Last edited:
#440
Intermittent

After removing the cover and measuring the key voltages, Vcc = 41, then 24, 12.16 and 5.1. Also tested the insulation of the transistors from the alum. sink which were ok. I tried it again and it started working.

So I don't know what it is yet, although the hand soldering job seems a bit poor.
Anyway I reassembled and it is still working for now.

Burnt out 2 small signal transistors by trying Hfe scripts w/o identifying if they were PNP or NPN. So the overcurrent LED works as well...😀.
Auto tested a DUT to identify an NPN...works fine.. Did some HFe scripts on it...seems ok. Seems the Ic is measured in decimal mA....but volts are in V.
 
Status
Not open for further replies.
Top Bottom