New model D problems

[Quatschmacher]

Got my new D week. It sounds lovely and looks beautiful. Sadly there were a number of issues which have necessitated its return. Mine had the following faults:

Mod wheel bent over to the right;
LFO knob off-axis;
Lower keys spongy (known problem with too-short aftertouch strip impeding key travel);
Other keys not always triggering correctly (eg 2nd highest Bb and B keys, lowest Ab, Gb and F too).

The other thing to note was the oscillator modulation switch throwing the pitch sharp by nearly a semitone when engaged when the mod wheel is fully up. (Can be verified as technically the off portion of the square wave LFO’s duty cycle and the sustain stage of the filter EG when set to zero should have no influence of the pitch as they should be outputting 0V at that point.) This is seemingly something that can be rectified - the aftermarket MUSE kit for the 2016 reissue solved this. Couldn’t Moog therefore also solve it?

Extremely disappointing, especially at this price point. There does seem to be quite a significant number of defective Model D units getting into buyers’ hands at the moment. This is disappointing for buyers, damaging for Moog’s reputation and must be quite costly at their end to end up having to deal with the returns, as well as dealers having to pay for return shipping. Surely it would be in everybody’s best interest to get these things right before the units leave the factory. A bit more time testing them rigorously before they go out would save everybody hassle in the longer term.

I’ve written directly to Moog to relay these points and hope they take some positive steps to put this right. As much as I love the instrument, I won’t be sinking £5k into something that isn’t functioning properly.

[VCO]

There does seem to be issues with Moogs quality control. I would request an exchange for one that is working properly.
Unfortunately Moog has a reputation for issues with thier synths . I had issues with my voyager. Hope you can remedy
a solution and it gets resolved. I would think a premium brand like Moog would want to resolve this issue not only for the sake of thier customers but for Moogs reputation as well.Hope things work out for you.

[sdwillingham]

Sorry to hear about your issues, Quatschmacher. I’m sure that must be very disappointing. I am lucky that my 2022 seems to be (almost) flawless. Hopefully, you can get your issues resolved soon. I have fallen in love with my Mini, and hope that you can have the same experience.

I definitely agree that Moog QC should be more thorough. Issues happen, but at this price, most of them should be caught and remedied before the product leaves the factory.

To be fair on both sides, I would like to clarify a couple of your points — mainly so everyone’s expectations are accurate.

1. The keyboard issues: It is hard to assess the cause of this issue. There might be a bad batch of Fatars, but it is also possible to be a chassis-interference issue that pops up in shipping. Hopefully, Moog will get to the bottom of it. But, I’m am almost sure the spongy F-key is not a “known issue” related to the design of the aftertouch ribbon being too short. The left-hand controller in the 2022 re-issue has been totally re-designed, and the AT ribbon is generously long:

2) The offset in the modulation switch is not a production issue, but is just the nature of the design. Moog added the option of routing the filter contour to modulation in the reissues. This is done in a simple manner that doesn’t modify the contour circuit. The sustain level will not generally settle to exactly 0V. Of course, the designers could have added some circuits and trims to make this work, but it may still turn out to be unstable or drifty without more extensive design. As is, the feature is useful for kick-drum patches and the like. It is very cool, but will never be precise, nor was it meant to be (IMO). As for the LFO, the offset is also a design choice — the LFO was designed purposely to be bipolar (just like OSC3). You characterize a square wave as having an “off portion”. However, in the analog world, just as many square-waves have positive and negative swings, for example +1 V and -1 V. In the Mini, the expected square-wave applied as modulation is two-pitches above and below the keyed pitch. It could be designed otherwise, but it is not a production or QC issue.

Again, hope you can get your Model D issues resolved soon.

[sdwillingham]

Quatschmacher, I want to apologize! In trying to “clarify”, I’ve muddied the waters a bit with some wrong information. I’ve gone back to examine my Mini more carefully.

I was completely wrong about the LFO & OSC3 square waves being bipolar! They do, in fact, swing from near zero to positive (after the modulation mixer — the swings are negative before). But, as you know, the near-zero part is also about 1 semi-tone sharp on my Mini (mod wheel full forward). Choosing the filter EG as modulator is slightly worse, but similar.

As for the reasons: Well, OSC3, the filter EG, and the modulation mixer are all vintage-design with discrete transistor circuits. Such circuits have limited gain and large DC offsets compared to modern opamps. The offsets still seem a bit bigger than I can account for, but I’ll study it further. I’m pretty sure the vintage Minis will also have similar offsets.

However, this is not the only cause of offset. The new left-hand controller board in the 2022 Model D has a completely different modulation path, implementing the ability to transmit and receive modulation via MIDI. This has two implications with respect to pitch vs oscillator modulation. First, there is a very small offset, even with the mod wheel fully off. This is due to a couple of opamps in the modulation path with millivolt-level offsets. On my Model-D, this causes about a 10-cent shift in pitch when the oscillator modulation switch is engaged.

The vintage Model-D, if the pitch-wheel is perfectly adjusted, would not have offset at the minimum-wheel setting. Also, a vintage Model-D at maximum modulation, will tend to cut the offsets of the prior modulation mixer by a factor of about two. (The 2016 reissue probably does this too.) The 2022 modulation path in the left-hand controller also reduces the prior offsets, but unfortunately, introduces a bit of offset of its own at full-wheel settings.

As I said before, this behavior is not ideal, but is in the nature of the vintage design constraints. For the most part, it will not significantly hinder many use cases, but I want to be as accurate as possible about the differences between the 2022 and prior models.