My implementation of the Cordell Distortion Analyser

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I am posting this in the Solid State Forum as there used to be a couple of permanent threads for Bob Cordell. So I thought it is more appropriate here than elsewhere.

A couple of years ago I was attracted to Bob’s DIY distortion analyser by a htread at the Pass forum :

http://www.diyaudio.com/forums/pass-labs/91745-distortion-analyzer-recomendations-2.html#post1335073

Someone has actually made one :

http://www.diyaudio.com/forums/pass-labs/91745-distortion-analyzer-recomendations-4.html#post1356316

But that link no longer works. So I could perhaps post the photos here without getting into trouble.


Patrick
 

Attachments

  • Cordell THD 1a.jpg
    Cordell THD 1a.jpg
    90.6 KB · Views: 5,212
  • Cordell THD 1b.jpg
    Cordell THD 1b.jpg
    99.4 KB · Views: 4,994
I always wanted one, but they were too expensive here even 2nd hand. An HP339 typically costs more than 400 Euros, and it also does not have balanced IN/OUT, which I want. Plus no service manual, probably many obsolete parts with no spares, …., that gave me enough excuse to build one myself.

As always with DIY projects, it cost just the same as a bought one in the end if not more, was a LOT of work, but also quite a bit of fun. Would have been more if it would produce music as well.

So here I am sharing with you some pictures of the finished article.
 

Attachments

  • DIYA Front.jpg
    DIYA Front.jpg
    119.4 KB · Views: 4,855
  • DIYA Overview Front.jpg
    DIYA Overview Front.jpg
    187 KB · Views: 4,616
  • DIYA Overview Rear.jpg
    DIYA Overview Rear.jpg
    506.4 KB · Views: 4,336
AX tech editor
Joined 2002
Paid Member
Another attempt was posted here, but I did not manage to find pictures of the finished article :

Circuits Online - Forum - Show your projects! Part 6.


P.

Nice! Brings back memories from when I build this unit, many, many moons ago. Alas, I no longer have this unit, but it worked flawlessly. A great design by a great designer!

At the time, I got the boards from Bob. BTW, where did you get the case? Looks quite professional.

jan didden
 
We are particularly proud of the battery case, as the thinnest wall behind the battery holder is only 1.5mm thin, and it is solid wood, not ply.

We do, however, want to redo the wooden end plates and the battery case, as the wood grains do not show up very well after dying them black. Minor cosmetic issues.


P.
 

Attachments

  • DIYA Batt Case.JPG
    DIYA Batt Case.JPG
    152.6 KB · Views: 1,298
Jan,

> Alas, I no longer have this unit, but it worked flawlessly. A great design by a great designer!

Could not agree more.

> BTW, where did you get the case?

It was a length of electrical trunking rescued from the scrapyard. Did making wiring a pain because of restricted access from the top, but saved me 100 Euros for a commercial enclosure. And for sure one of a kind. This particular extrusion that I have is no longer available, but this is something similar :

OBO BETTERMANN - LFS. GEK-A device installation trunking aluminium systems, 80 mm cover

The end plates and the battery case were milled from solid beech wood (Buche Leimholz).

The lettering was done by Lazertran :

Lazertran Waterslide Decal transfer paper for use with photocopiers

And I did use the baked technique.


Patrick
 
Last edited:
I wouldn't bother today - add a few filters to supress fundamental and use a sound card - indirect IMD measurement resolution can reach way below human hearing noise floor with averaging

the ESI Juli@ is ~US$120 and has great resolution and low distortion by itself
 
www.hifisonix.com
Joined 2003
Paid Member
JCX, is the problem with a sound card not the fact that the bandwidth is limited to c. 22kHz? How do you do a wide bandwidth measurement (e.g. 80Khz like the AP)? Another thing is frequency response (although not covered by Bob's instrument) - sound cards do a great job of measuring 5Hz to 22KHz - but after that what do you do?
 
Great job- mine isn't as pretty!

I built my Cordell analyzer over a period of a couple years, starting right after the Audio magazine article came out. Etched my own boards from the published patterns. It's a super instrument and I wouldn't trade it for the world. Sound cards are great as far as they go, but bandwidth, calibration and certainty of the result are all issues. You can get wildly different results depending on sample rates and bit depths. The only way I'm sure of what my sound card is doing is to compare it with a conventional THD analyzer. Also, at least with my Emu, the ADC quality is great, but the DAC snr is unimpressive. One would never mistake it for a high quality analog signal generator. The noise isn't obvious in a spectral plot, but sticks out like a sore thumb on a scope.

The only downside to building a Cordell these days is the expense and difficulty of building up the switches. Sometime I'll post a photo of mine. It's been trouble free and the floor is about 0.0002% midband.

Best,
Conrad
 
I appreciate fine analog circuitry and good dedicated function test equipment as much as anyone - but the lazy cheapskate in me can't ignore the possibilities of today's good&cheap digital hardware

the ESI Juli@ has up to 192K sample rate so it can make measurements comparable to the classic 80 KHz BW audio distortion analyzers

with a few active filter circuits to cut the fundamentals of the test signals and boost the error signal you can resolve with very good accuracy with better soundcards - spurious free dynamic range before the added filters can be ~100 dB, indirect measurement tricks with two channels generating IMD test tones and added gain/filtering can allow resolution of nonlinear test circuit behavior to exceed 160 dB

Free SciLab or other processing software can extract anything you can write the equations for from the .wav files
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.