Little design flaw in the ODR-1 ?

[Vivek]

I know of mathematicians scribbling unproven conjectures and notes on the margins of books, and other mathematicians taking 200 years to prove the hastily written scribble.

Similarly, ACDC once casually mentioned that they use their wireless system even in the studio where the do not really need wireless, because it adds something desirable to their sound. That lead to the quest to find out what exactly was in the wireless system that enhanced their sound.

Now at https://nordland-electronics.de/en/products/odr-c.html, Kai Tachibana slyly lets slip :

"There have been quite a few companies that have rebuilt (copy) the ODR-1. As there were Wampler, Visual Sound, Rocket, Vemuram and maybe others. Of course, always with a "small" change, so it is not a 1:1 copy. Funnily, none of these "engineers" really noticed my little design flaw from 1992 in the device. He was diligently copied. But I do not say what that is. Treat me to my fun!"


So wise brothers, please check the ODR-1 schematic at https://github.com/yuvadm/guitar-effects-schematics/blob/master/Distortion%20Boost%20and%20Overdrive/Nobels%20ODR-1%20Plus.pdf

or https://schematicheaven.net/effects/nobels_odr1_overdrive.pdf

and let's collectively detect the elusive "little design flaw".

[wampcat1]

I think Kai is misunderstanding the game a bit, or perhaps I'm misunderstanding what he's saying here... There's potential "flaws" in many circuits, but when cloning a circuit a company will often keep those sorts of things in, as general customers tend to believe that there is some magic in each circuit. Take the Klon circuit for example, it's a bit silly and the same result could be had much easier, but that's not the market.

[Mark Hammer]

Markets are funny things; sometimes spurred by fads, and other times spurred by superstition as much as anything else.

But while I've got you here Brian, a big old Hiya from up north.  Nice to see you.

[Fancy Lime]

Well, the ODR1 contains a number of design choices that seem peculiar if one only considers the schematic. Unnecessary resistors and caps, odd values, etc. But some of those may have to do with parts availability or cost or with layout convenience. Really hard to pass judgement on these things and almost all products ever produced are to some degree a bit silly like that. Especially stomp boxes from the early days, when a lot of later legendary designers didn't really know what they were doing yet, and struck gold by accident. Consider the Rat.

The only thing that leaps to my eye that I would categorize as a flaw instead of a quirk, is the position and value o R11. That should probably be place before R10 and be 100k, so as not to reduce the signal before the gain stage. This would improve the signal to noise ratio slightly. But it would also change the gain structure ever so slightly, so this thing that I would consider a flaw may well be a deliberate design choice on the part of the original designer.

Anyway, Brian's answer nicely demonstrates the stereotypical difference between American and German design priorities. American engineers design to please the customer. German engineers design to impress other engineers. That's why American cars are successful and German cars don't break. Not sure the that is indeed universally true, but that's what an American engineering student told me decades ago after an internship at a German car manufacturer. May well be just a marketing cliché.

Andy

[r080]

For comparison:
https://aionfx.com/app/files/schematics/nordland-odrc-trace-schematic.png

[lcv]

For U2A , both inverting and non-inverting inputs seem unnecessarily  DC biased , through  R27 and R24, creating a potential small output offset ?
R27 should go  to V/2 via a cap.
Just my  2 cents

[ElectricDruid]

Is this the same as this?:

https://aionfx.com/app/files/docs/gale_documentation.pdf

(Wireless system used by AC/DC in the studio, etc etc - here without the wireless function, for just the "toan").

[anotherjim]

I think that not having an input cap is a flaw. I know the JFET doesn't need one to work, but if you plug it in after something with DC leakage it's this box that will be perceived as the culprit.
Could also complain about the unity gain output buffer without a series resistor to protect from the capacitive load - that seems to be corrected in the Aion version.

Speaking of slavish copying, re the Gale - does the LM386 need a Zobel network on its output in this case?

[antonis]

Speaking of slavish copying, re the Gale - does the LM386 need a Zobel network on its output in this case?

IMHO, No..!!

Can't imagine that particular output configuration to be directly connected on a speaker (or any other inductive item)..
(maybe it should be needed if Volume pot was placed back in pin 3 and C11 value raised X20 to X40..)

[Ben N]

I think that not having an input cap is a flaw. I know the JFET doesn't need one to work, but if you plug it in after something with DC leakage it's this box that will be perceived as the culprit.

Does this really matter, as there is a coupling cap off the source of the FET blocking any DC downstream?

[Güero]

design flaw from 1992 = Klon mysterious diodes

[r080]

For U2A , both inverting and non-inverting inputs seem unnecessarily  DC biased , through  R27 and R24, creating a potential small output offset ?
R27 should go  to V/2 via a cap.
Just my  2 cents

R27 goes to ground in the dec 2000 schematic. This might need an earlier version schematic to figure it out.

I don't know why I didn't get it before, but I only just now realized how the ODRC's ODC control works.

[Rob Strand]

"There have been quite a few companies that have rebuilt (copy) the ODR-1. As there were Wampler, Visual Sound, Rocket, Vemuram and maybe others. Of course, always with a "small" change, so it is not a 1:1 copy. Funnily, none of these "engineers" really noticed my little design flaw from 1992 in the device. He was diligently copied. But I do not say what that is. Treat me to my fun!"

The way I'm reading that is, it is expected the "copies" will have small intended differences but the copies managed to copy the fault while "He was diligently copied."   

I thought the flaw jim picked up made the most sense.   If my interpretation of the quote is correct then we need some of the copies to confirm they have the fault.   I don't know what the copies are precisely and hence I haven't looked for schematics.   Only when those are found can we see what flaw has been copied.

As already mentioned there are many possible design decisions and what someone might call a flaw would still be "ok".  I have a feeling it's not this type of thing he's talking about.

Another angle is he himself copied the circuit from some other design and he made an error while copying it.   Without knowing what that base design was there's no way to see what part he stuffed up.

Even harder to detect is he made an error which didn't follow his original intentions.   That's just about impossible to detect.  For example what if he accidentally flipped R12 and R13 or C12 and C13.   It's impossible to read someone's mind to detect such a part flip.

The flaw jim picked up is an actual flaw and it doesn't need any surrounding circumstances to explain it.

[Rob Strand]

R27 goes to ground in the dec 2000 schematic. This might need an earlier version schematic to figure it out.

I think we need the early schematic also.  I'm assuming the 1992 version with the flaw is the version with the Tone control and not the Spectrum?

As far as the input cap on the JFET is concerned:

https://www.nordland-electronics.de/en/blog/odr-1.html

"Input

For (almost) all effect pedals, I used a JFet circuit as an impedance converter at the input of the circuit. Unlike some other manufacturers, I put a lot of emphasis on a version without an input capacitor, just like typical tube amplifiers. I could also have taken the high-impedance plus input of the first opamp - just like many other companies did for cost reasons. But after long tests the ear decided in favor of the JFet and in spite of the additional expenditure of material and assembly work I stayed with it."

In his eyes the lack of cap could hardly be perceived as a design flaw!

[antonis]

The whole story reminds me a case of a Greek geographical company, claiming for some intentionally made flaws for preventing contraband copies of their maps..
(the truth was that those "intentionally" made flaws were actually due to their own oversight/slopiness during conture plan mapping precedures..)

[Rob Strand]

The whole story reminds me a case of a Greek geographical company, claiming for some intentionally made flaws for preventing contraband copies of their maps..
(the truth was that those "intentionally" made flaws were actually due to their own oversight/slopiness during conture plan mapping precedures..)

The same company made the maps for Atlantis and look what happened to that.

[anotherjim]

I think that not having an input cap is a flaw. I know the JFET doesn't need one to work, but if you plug it in after something with DC leakage it's this box that will be perceived as the culprit.

Does this really matter, as there is a coupling cap off the source of the FET blocking any DC downstream?

A positive voltage on the gate reduces -Vgs. The Fet biases hard-on.

[Ben N]

Ah, I kinda suspected it might be a biasing issue.

[Rob Strand]

OK, lets start with the wampler copy.

Is this it?  It's the Deluxe Paisley.  It's a dual overdrive.  One one half is the TS9-ish Paisley and the other the ODR-S-ish Underdog.  No schematic found for the whole thing but there is a Paisley schematic.

From,
https://www.wamplerpedals.com/blog/latest-news/2017/10/rooting-for-the-underdog/

"Released in 2009, with a limited run of just over 100, the UnderDog is based on a Nobels ODR-S but modified to clear up some of the shortcomings Brian felt were inherent in the design. "

Note it says ODR-S.
https://schematicheaven.net/effects/nobels_odrs_overdrivespecial.pdf

This schematic does in fact have a design flaw: U1B pin 6 is getting its input bias through the clip diodes!

The limited production stand-alone Underdog is this one,
https://www.effectsdatabase.com/model/wampler/underdog

I can't find a schematic for the Underdog however the on-line Paisley schematic and layout has a JFET input buffer *with* an input cap.   So if the UnderDog follows the Paisley the lack of input cap isn't the "Little design flaw in the ODR-1" because that has not been copied.   The Paisley drive looks more like a TS9, no clip diodes to ground.  What about the Underdog? I'm expecting clip diodes, but is the opamp bias flaw copied as well?

(BTW, the flaw is not on the ODR-1 (2000) schematic.  R24 diverts pretty much all the bias current from the diodes.)

[Vivek]

Note it says ODR-S.
https://schematicheaven.net/effects/nobels_odrs_overdrivespecial.pdf

This schematic does in fact have a design flaw: U1B pin 6 is getting its input bias through the clip diodes!

Thanks Rob,

What really does not perform well due to oddball scheme to bias the + Pin via the clipping diodes ?

Is this a "Flaw" as in something does not function well

or a "Quirk" as in a funny way to do things ?