Subscribe to EDN

Audio Precision to the rescue, of Jim Williams no less

March 29, 2011

Chris Gill, the Audio Precision sales engineer serving Silicon Valley was nice enough to bring their latest and greatest over to Linear Technology where he took some measurements of a Jim Williams’ low-noise/ low distortion oscillator. Chris was really cool, no sales pitch or pressure to buy one, just pure customer service goodness. It was great since he really knew the SYS-2722 and was able to get good data in a few minutes, using his own laptop.

chris_gill_jim_williams

  • Here is Chris and Linear Technology’s Jim Williams, setting up the SYS-2722 to measure a 1.8kHz oscillator Jim designed.

chris_gill_jim_williams_data

  • Here they are pouring over the data like any two great engineers would.

sys-2722_setup

  • This is the setup; Jim Williams’ oscillator board is at bottom left.

audio_precision_usb_20.jpg

  • The SYS-2722 comes with this handy USB adapter, I am pretty sure Chris told me it was included in the price.

Jim Williams had to design this ultra-pure low-distortion oscillator to use in a circuit feeding a Linear Technology high-performance ADC. After all, how do you know how good your ADC is working if you feed it a crappy sine wave full of distortion and artifacts? So Jim did his first crack at the circuit a few months ago. He sent it to Portland, to Bruce E. Hofer, the former Tektronix engineer and founder of Audio Precision. Bruce measured the oscillator and sent it back along with the test data. You gotta love a guy whose title is “Chairman, co-founder, and head analog engineer”.

Well, Jim Williams is the kind of engineer that seeks to improve things. So he did some tweaks and redesign on his oscillator circuit. Jim didn’t want to bother Bruce again; he is the chairman of AP, after all. So Jim asked me if I could scrounge up an AP box we could use to measure the SNR and distortion. My buddy Joe Curcio, an apps engineer at National Semiconductor owns the classic Audio Precision System-1 (S1), but wondered if we couldn’t get our hands on some brand-new calibrated hardware.

That’s when AP vice president of marketing Tom Williams put me in touch with Chris Gill. Shortly thereafter Chris met me over at Linear Tech and we saw the several dB improvement Jim had made in his oscillator board. It proves my point that the key to being a good design engineer is knowing how to test, and having the equipment to test your designs.

I don’t want to steal Jim’s thunder, we are going to publish his whole design and the test results in our August issue. Still here is a little crop, just the corner of one of the beautiful plots Chris made with the SYS-2722. Even Jim was impressed. The SYS-2722 has a quieter, lower-noise oscillator built in, and that one sweeps across the audio band, rather than a fixed 1.8kHz like Jim’s.

sys-2722.gif

  • I don’t want to spoil Jim Williams’ article by giving away the data a few months early, but lets just say the axis I cropped off on this Audio Precision SYS-2722 measurement goes down to -160dB.

Audio Precision has really built up a great brand. I have dozens of inputs that tell me this. When I was at National Semi, all the engineers in the audio group used to talk about how great AP stuff was, and this was 10 years ago. My electronic flea market buddies would be talking about how great the AP stuff was and eventually Joe Curcio bought a used unit to have at his house. Then, when I visited Tektronix up in Portland and mentioned my next stop was Audio Precision, one of the engineers sighed and said, “We never should have let Bruce leave Tektronix!” Finally, I met with the Audio Precision folks in person. When marketing VP Tom Williams took me over to Bruce Hofer’s cubicle, there was an Audio Precision product half disassembled on his desk surrounded by a scopes and meters. That’s the kind of chairman I want to work for.

audio_precision_team

  • Here is the Audio Precision team, that’s Bruce in the middle holding up the sign. He always wants to pitch in and help, just like any great boss.
Posted by Paul Rako on March 29, 2011 | Comments (12)

April 13, 2011
In response to: Audio Precision to the rescue, of Jim Williams no less
Jim Williams commented:

My trusty Audio Precision System One A rescues me every day here. I'm suprised more silicon valley dudes don't have them. I found for sensitive measurements I need to power off all the bench gear and never use a CRT monitor, that will skew your measurements. Florescent lighting should also be avoided. AP gear is so good you learn quickly the limitations of your enviroment.


April 12, 2011
In response to: Audio Precision to the rescue, of Jim Williams no less
jcdrisc commented:

I know Jim's low distortion osc designs from the LT appnotes. I have an old HP200C I want to reuse, I hope Jim has air capacitors in the wien network !!


April 8, 2011
In response to: Audio Precision to the rescue, of Jim Williams no less
sheryl commented:

Ronnie,
Try the "Data Converter Test Module" from LTX-Credence.
they gave a paper at this past year's ITC with impressive results.


April 6, 2011
In response to: Audio Precision to the rescue, of Jim Williams no less
Ronnie Yujuico commented:

Hello Bob, using an analog scope in crossplot mode to measure distortion also requires a very low distortion signal generator; may I know what signal gen you are using to achieve less than -150 dB?


April 1, 2011
In response to: Audio Precision to the rescue, of Jim Williams no less
rich l commented:

bob, do you have any info regarding the common mode distortion for these amplifiers? That seems to be what is limiting sfdr in my app. also the output stage into cap load.
Why don't mfg specify common mode distortion? In my app the limiting fact to get beyond -150db the at any frequent or amplitude ist the amplifier not the converter.


March 31, 2011
In response to: Audio Precision to the rescue, of Jim Williams no less
Bob Pease commented:

I have been measuring op-amps with -159 dB of distortion, using an old Tek 2445 ANALOG Scope in Crossplot mode. I do this on the bench in my office because I don't have any A.P. equipment. I'm pushing past -164 dB next. / rap


March 31, 2011
In response to: Audio Precision to the rescue, of Jim Williams no less
Al C commented:

I would expect the principal spur to be a third harmonic. Are the 5 or 6 digit frequency readouts only good to three digits?


March 31, 2011
In response to: Audio Precision to the rescue, of Jim Williams no less
There is always one... commented:

Hey! "Rich I"; if you've done such a great job, why not post YOUR DESIGN? As it says in the article, Jim's design will be posted in August for everyone to see, and from the looks of it with some damn fine testing and results. Are YOU up to doing the same? Put your money where your mouth is!


March 30, 2011
In response to: Audio Precision to the rescue, of Jim Williams no less
rich l commented:

bogus! don't believe what u see!
I know, I have developed a system w/better than -140db sfdr AP couldn't do it. Get real


March 30, 2011
In response to: Audio Precision to the rescue, of Jim Williams no less
Pesky Varmint commented:

How many clueless MBA's, finance people, and CEO's will never figure out why and how AP is successful.


March 30, 2011
In response to: Audio Precision to the rescue, of Jim Williams no less
Brad commented:

Yes, Bruce is definitely one of the greats, and yes, Tek was insane not to persuade him to stay on board. And they began a significant downward spiral, which bottomed out with some truly terrible 'scopes --- one of which I had to use for a while, and which actually lacked the ability to trigger ch A sweeps from ch B!!
Tek has recovered, fortunately, but doesn't dominate the 'scope market as much as they used to.


March 29, 2011
In response to: Audio Precision to the rescue, of Jim Williams no less
Andy T commented:

"Here they are pouring [sic] over the data like any two great engineers would."
NAH!...they are just a pair of experienced and OLD farts who forgot their bifocals at their desks
:-)
The big challenge, of course, would be to see if they can talk Pease into using that LCD display/typewriter-looking thingy that's on top of the meaurement instrument.

POST A COMMENT
Display Name
captcha

Before submitting this form, please type the characters displayed above. Note the letters are case sensitive:

Advertisement
Advertisement
Advertisement
About EDN   |   Site Map   |   Contact Us   |   Subscription   |   RSS
© 2012 UBM Electronics. All rights reserved.
Use of this Web site is subject to its Terms of Use | Privacy Policy

Please visit these other UBM Canon sites

UBM Canon | Design News | Test & Measurement World | Packaging Digest | EDN | Qmed | Pharmalive | Appliance Magazine | Plastics Today | Powder Bulk Solids | Canon Trade Shows