Little design flaw in the ODR-1 ?

[antonis]

IMHO, just an asymmetrical clipping offset of a diode forward voltage drop..

[bean]

Perhaps shorting C20 so there is about equal amounts of vRef on both sides on the clippers? Alternatively, referring to the ODR-1 vs ODR-S schem, the ODR-1 has a 43k pull-up to vRef.

[Rob Strand]

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 ?

I've got my doubts it would be audible in practice.   It's more of an undesirable thing to do in a design.

Notice the ODR-S has germanium diodes and the ODR-1 has silicon.

The opamp input bias current is in the order of 5nA to 200nA, upper end unlikely.   That's very low.
With such a low bias current the voltage drop of a silicon diode is only 60mV to 220mV.
It's not going to take much signal to overcome that.  The diodes are fed by 10k (or 12k) so 5nA to 200nA
is in the order of 50uV to 2mV drop across 10k.   The other way to look at it is at 5nA to 200nA
the dynamic diode resistance is going to be 220k to 9M so it's not going to do much against
a 10k resistor.   A clipper at these levels is going to sound quite clean to the ear although you might be able
to measure some distortion with test equipment.  Even at the extreme of 220mV offset the loss in swing
at the opamp isn't going to matter since the clippers prevent the opamp from clipping.

An interesting test would be to deliberately put some DC bias current through a clipper fed by 10k.  Pass a signal through the clipper that is low enough in level that it still sounds clean.  At what current do we start to hear the signal
level drop and at what level does the signal become distorted (if at all).

FWIW, we still don't know if the DC bias through the diode is the "design flaw" the designer was referring to.  That's my best guess so far after pooling the (cough) "evidence".

[aion]

A bit late to the party, but earlier someone posted my trace of the ODR-C, which is Kai's updated/expanded/boutique version. I have a pretty extensive write-up about it here:

https://aionfx.com/news/tracing-journal-nordland-odr-c-custom-overdrive/

In particular, I bulleted out every difference between the ODR-C and the ODR-1 factory schematic.

My understanding is there's some ambiguity as to whether he's saying there's a mistake in the factory schematic or a mistake in the actual pedal itself, but going through those bullet points might be helpful, scoping out whether any of the differences seem to be corrective in nature as opposed to improving, optimizing or adding.

It's definitely not the lack of an input cap, though, since both the ODR-1 and ODR-C are the same in this regard and he's talked about this design choice specifically.

[Rob Strand]

I have a pretty extensive write-up about it here:

https://aionfx.com/news/tracing-journal-nordland-odr-c-custom-overdrive/

In particular, I bulleted out every difference between the ODR-C and the ODR-1 factory schematic.

I like your traces. You always do a thorough job. Nice breakdown as well.

As far as the "design flaw" goes, I'm seeing this progression:
- ODR-S   clip diodes pass full opamp bias current.  In fact only one diode passes the current.
             (did the early ODR-1 do the same?)
- ODR-1   43k added (R24 on original ODR-1 (2000) schematic, same as R22 on ArionFX ODR-C schematic)
              Resistor bypasses opamp bias current away from diodes.   Insignificant current through diodes remain.
- ODR-C   C12 added
              Prevents opamp bias current from passing through diodes altogether.

The thing that stands out from your write-up is:
- ODR-C: A 470R resistor (R41) has been added in series with the output.

I'd call that a design flaw.   It's good design practice to have such a resistor as it prevents opamps and buffers oscillating with capacitive loads, such as a long cable.  Plenty of pedals don't have it but it's a very risky practice.

My understanding is there's some ambiguity as to whether he's saying there's a mistake in the factory schematic or a mistake in the actual pedal itself, but going through those bullet points might be helpful, scoping out whether any of the differences seem to be corrective in nature as opposed to improving, optimizing or adding.

Interesting, that adds yet another perspective to list I posted before.

I'm not 100% sure why he added all the 10 ohm resistors.  You see things like this in equipment using single layer boards.  They are effectively used as 0R links as 0R weren't so common.  Using  a resistor stream-lined the manufacturing process compared to wire links.   In the pedal they may help with oscillation and/or interference from mobile phones (although there's probably better places to put them).

[Rob Strand]

The thread might be about flaws but the pedal sounds pretty darn good.

Also pretty good for 1991/1992 when the TS-9 was the main sound.
Funny thing is I also was building transparent pedals back then but thinking
back I like the ODR-1 more, so a big thumbs up to Kai Tachibana.

Nice review/run down here,

What About The Nobels ODR-1? [vs TS9 & BD-2w into AC15 & Deluxe Reverb] That Pedal Show

[Ben N]

The thread might be about flaws but the pedal sounds pretty darn good.

I built one on Kevin's board and I love it to death. I built it with the bass control option, and I had it rolled down somewhat at first because I was playing into a Vibrochamp and had to limit bass so I didn't overdrive the speaker. Once I went back to a 12" speaker, I kept the bass rolled off a bit, but thought it was sounding a bit anemic. After hearing the TPS demo, I turned the bass back to 100% on (don't know why it took so long to try this), I was hearing the full roar of this thing again. Point being, the bass is a big part of the sound of this pedal. It's really, really good, and if you need to tighten it up for leads, just stack with a mid-humper.

[Vivek]

I began this thread when I had started to analyse this circuit using LTSPICE

Then life happened to me while I was busy making other plans.

But this thread resurrection has now motivated me to complete the SPICE analysis.

[iainpunk]

didn't re-tracings of the circuit show a 10Ω or 100Ω resistor in series with the clipping diodes? seems missing in the original circuit?

cheers

[Rob Strand]

I turned the bass back to 100% on (don't know why it took so long to try this), I was hearing the full roar of this thing again. Point being, the bass is a big part of the sound of this pedal.

The amount of bass boost is really tough to get right.   With light boosts you can use a quite a bit of bass boost and it sounds nice and flat.  When you start cranking the gain it becomes indistinct and woofy.    I found as I would tweak the pedals over time there was a progression to back-off the bass and then you end-up with the mid-hump again and it loses the flatness.  (After that I tried different bass boost schemes.)

didn't re-tracings of the circuit show a 10Ω or 100Ω resistor in series with the clipping diodes? seems missing in the original circuit?

There's a few variants, which I listed a few posts back.  Some have the 10/100 ohms some don't.  The official ODR-1 schematic from 2000 does not.   As already mentioned in the thread it's not clear what these resistors are doing.  They could be used as PCB links, or they could be to help EMC issues, or they could be used as place keepers for future mods. [FYI, the ODR-C schematic has some extra resistors.]

[GGBB]

I built one on Kevin's board and I love it to death. I built it with the bass control option, and I had it rolled down somewhat at first because I was playing into a Vibrochamp and had to limit bass so I didn't overdrive the speaker. Once I went back to a 12" speaker, I kept the bass rolled off a bit, but thought it was sounding a bit anemic. After hearing the TPS demo, I turned the bass back to 100% on (don't know why it took so long to try this), I was hearing the full roar of this thing again. Point being, the bass is a big part of the sound of this pedal. It's really, really good, and if you need to tighten it up for leads, just stack with a mid-humper.

The bass control in the Aion Andromeda lowers gain as well as raises the bass filter cutoff frequency. A better approach to reducing bass might be to leave gain alone and simply raise the cutoff by using a smaller cap than the 2µ2, since the gain change actually affects all passed frequencies a bit. I suspect - but don't know for sure - that's what the bass switch in the ODR-1 BC does. In my own clone I used a 1µ and I think it sounds great that way - still very "flat/transparent" but not at all bassy.