Klavis - CalTrans

[bgribble]

If I then go through a neutrally calibrated I/O of CalTrans, it's off by a lot.

One thing that is tricky about the CalTrans is getting your “octave staging” right. Try switching your input voltage up an octave and your oscillator tuning down an octave or vice versa. Note that V/oct inputs are positive only and Klavis has said that accuracy is not great near 0v input.

Also are you sure that you have the channel under test neutrally calibrated?

[skxdlxx]

If I then go through a neutrally calibrated I/O of CalTrans, it's off by a lot.

One thing that is tricky about the CalTrans is getting your “octave staging” right.

Ah, haven’t thought about that, will give it a try. Definitely on a neutrally calibrated channel. From my understanding the neutral mode is just a 1:1 of what goes in is what comes out - perhaps I’m reading that wrong somewhere

[donutschool]

I'm revisiting my CalTrans and still struggling with 'neutral calibration'. If Im connected directly to an oscillator, my tracking is good. If I then go through a neutrally calibrated I/O of CalTrans, it's off by a lot. Perhaps its a beginners modular problem im having

(Assuming you're on 1.17 firmware and have the latest V1.5 manual...)

What happens if you do a V/Oct calibration (p9 of manual) - does the tracking work then? If so - and you calibrate it to an oscillator that already has good tracking - then the result won't be much different from neutral calibration anyway.

If it still doesn't track then the "Calibrating the inputs scaling" (sic) on p16, using a tuner +,VCO*, should put it right. If you trust the tracking on your VCO,that boils down to alternately playing 2 notes (say) 3 octaves apart, fine-tuning your VCO to match the first note then follow steps 9-10 from p17 to bring the second note into tune. Rinse and repeat, and maybe try reversing the two notes (tune the VCO to the second, adjust Caltrans to the first) if you have problems.

Also, if you're trying to check/set calibrations, I'd avoid using notes at the extreme ends of either the Caltrans' input range (0V - 7V) or your MIDI controller's CV output range, which sometimes get a bit non-linear.

* Don't try and use the voltmeter method unless you've got a "professional" one with something like 0.05% or better accuracy on the 5=10V range - the manual is a bit misleading here - "4.5 digit display and millivolt precision" is irrelevant if the meter only has ~0.5% accuracy (which is a quarter of a semitone on a 4V v/oct signal!) - which is pretty common on handheld meters, even from good brands (your money gets you a more honest figure on the specification sheet). If you didn't pay several hundred bucks for the meter, it probably won't cut the mustard, and even if you did you need to check the specs.

[SunPulse]

The tracking on my system has always had a mind of it’s own, even over one octave things would get wonky. After reading about the CalTrans I ordered one from Thomann & received it yesterday. So here’s a little rundown on how I got along with installing, updating the firmware & calibrating my system.

I have a smaller eurorack in 2 Intellijel 84HP cases. In one is a Pittsburgh Modular SV-1 & in the other case are 2 WMD/SSF Spectrum oscillators. So 4 oscillators in total, which works out nice with the CalTrans having 4 outputs.

I use a 5-octave Novation keyboard which sends MIDI to the SV-1, the SV-1 then sends CV to the Spectrums through an Intellijel buffered multiple. I’d already adjusted the tracking on all the oscillators as best I could with a Korg rack mount tuner, but there was definitely room for improvement.

I installed the CalTrans in the same rack as the SV-1. The first thing was to update the firmware to v1.17. I downloaded the file to my iPad and used a lightening to 1/8” jack adapter to send the firmware update to the CalTrans. I set the volume on the iPad to just above halfway and the upload was successful on the first try. Good start!

I then followed the calibration procedure on page 9 of the manual for each oscillator. When done each oscillator was outputting a sharpened E when playing a C on the keyboard so I adjusted the semitone for each channel on the CalTrans to a C & then adjusted the fine tuning on each oscillator to get them tuned to C correctly.

When playing up and down the keyboard the oscillators were in staying in tune with each other, but they would all go equally sharp as I played up the keyboard. On page 11 of the manual it says if this happens to try changing the octaves on the CalTrans to determine if the CV voltage to the oscillators needed calibration. Switching octaves on the CalTrans the oscillators stayed perfectly in tune, so I had a new problem I didn’t realize that the CV output on the SV-1wasn’t tracking properly.

I then followed the calibration procedure on page 16 “Calibrating the Inputs Scaling”, and the oscillators are now tracking the keyboard really well and are staying in tune with each other. Not perfectly, but they are analogue and that’s part of being analogue. The important thing is they don’t require any fine tuning adjustment over the 5 octave range of the keyboard.

Overall I’m really impressed with how this all turned out. I read the manual carefully and made sure to do exactly what it said to do, and everything went correctly on the first try. I didn’t have to redo anything. The oscillators now track really well across the 5 octave range of the keyboard, which is more than I expected. I would have been happy with tracking over a 3 octave range! I can now rely not having to retune the oscillators depending I what notes I am playing.

So the CalTrans is an excellent module which does deliver what it promises. You do have to follow exactly what the manual says, none of the “I don’t need to read that” mindset when calibrating your system. Very reasonable price & quick delivery from Thomann to Canada. I wish I had a CalTrans 2 years ago!

[otoskope]

How does CalTrans go with oscillators such as metasonix RK7, buchla red panel 158, etc? Anyone tried?

Hi,
I've done quite some research into adaptive pitch control of Metasonix modules using continuous feedback and pitch tracking (implemented in Nord Modular and Nord Modular G2). But I also have a CalTrans on order, and will try it with Metasonix and other unconventional signal generators. I'll post the results here.
/Palle

[drowld]

How does CalTrans go with oscillators such as metasonix RK7, buchla red panel 158, etc? Anyone tried?

Hi,
I've done quite some research into adaptive pitch control of Metasonix modules using continuous feedback and pitch tracking (implemented in Nord Modular and Nord Modular G2). But I also have a CalTrans on order, and will try it with Metasonix and other unconventional signal generators. I'll post the results here.
/Palle

Any news ? i'm curious if it would work or not with the red panel 158

[Footkerchief]

One thing that is tricky about the CalTrans is getting your “octave staging” right. Try switching your input voltage up an octave and your oscillator tuning down an octave or vice versa. Note that V/oct inputs are positive only and Klavis has said that accuracy is not great near 0v input.

I've had a CalTrans (Yarns -> DPO) for a few weeks now, and this is dead on.

I'm very happy about the tracking it enables in the sweet spot. However, the tracking gets weird around the edges. For instance, it seems like the outputs are designed to drop down an octave when receiving a note above their range. That's a good fallback, but it would be great if the module also directly communicated that you're asking it to do something it can't do. Maybe flash an LED on channels that have out-of-range notes?

By the way, anyone know the min and max output voltage?

[Zentek]

I'm very happy about the tracking it enables in the sweet spot (using with a DPO). However, the tracking gets weird around the edges. For instance, it seems like the outputs are designed to drop down an octave when receiving a note above their range. That's a good fallback, but it would be great if the module also directly communicated that you're asking it to do something it can't do. Maybe flash an LED on channels that have out-of-range notes?

By the way, what's the min and max output voltage?

There is a reason for jumping octaves when asked for going beyond/below the range of a target VCO.
When two or more voices are transposed jointly (while playing in harmony), the one that cannot follow so high or low, will be brought back within its possible range while preserving the harmony of the whole. Otherwise the melody of those voices will be "clipped" and stuck at their farthest note.

Indicating such behavior as you suggest is a good idea. Unfortunately this should relate to each channel, whose LEDs are already busy with channel selection and quantizer blinks. I don't see how to add yet another visualization to these LEDs.

The I/O voltages and more are found on the last page of the manual.

Thank you for the appreciation.

[Footkerchief]

Thanks, I missed that last page of the manual! And good to know about the LEDs -- appreciate the reply.

[Ghost_the_garden]

Still no report on whether this works with metasonix modules? Wouldn't it be crazy if this thing could coax melodic lines out of an r-54!? I'd test if out if some one sent me the module~for research! (:

[otoskope]

Still no report on whether this works with metasonix modules? Wouldn't it be crazy if this thing could coax melodic lines out of an r-54!? I'd test if out if some one sent me the module~for research! (:

TLDR: Yes, you can (but it was a bit tricky)! Sorry, I sort of promised to get back about this. I posted a month ago about it in the Metasonix fb group:
https://www.facebook.com/groups/2183255 ... 998407820/

If you can't access it, here's the text part of my fb post, but I can't post the video here (may upload to youtube or something, but can't do that right now):

Klavis Caltrans as pitch CV controller for Metasonix oscillators
Playing Metasonix modules melodically has always been a challenge, and sometimes takes lots of calibration. Or you sync them with a pulse from another oscillator into the pitch input. I've been exerimenting with self-adaptive real-time patches in the Nord Modular, but that may not be available for everyone.. The small module maker Klavis has a 4ch trainable voltage mapping module (ca €150), and I bought one to try with Metasonix. Here's a test with R55.
During training, the Caltrans basically runs through the CV input range of an oscillator while listening to it, and it finds out how far it goes and which voltage corresponds to which pitch. You can hear that process in the beginning of the video. Sort of like the old Metasonix R60 module, but faster, cheaper, with four channels, and with presets, octave control, transposition and portamento/glissando. After training and setting trasnposition right (on the Caltrans octave and semi knobs), you feed it normal V/oct pitch CV, and it translates it to whatever needed.
It took some tries before I found the right settings on the R55 (gain & master tuning), but finally I was able to get a little bit more than 3 octaves of OK tuning, as demonstrated in the video. During training I have the overdrive set do 0dB, and I run the oscillator through an LPG to remove higher harmonics before it hits the Caltrans training input.
In the video, the CV and arpeggiation is generated from a Squarp Hermod (V/oct out, played from a MIDI keyboard).
I was able to get it working with the R54 too. Most likely it will work with RK4 and RK7, as they contain similar circuits, but I don't have them so I can't try.
The tuning isn't perfect, as these tube monsters are inherently unstable. But the Caltrans helps a long way towards playability. One minor error source may be that the pitch is vactrol-controlled inside the R-modules (?) and that the Caltrans maybe doesn't wait until the vactor fully stabilizes, Not sure.
Note that you can use one Caltrans on four oscillators simultaneously, as it has four individually trainable channels. And the whole setup can be saved in one of four presets. It can also be used to play in pitch on feedback circuits, and other weird patches. Anything that oscillates and where the pitch is controllable by some CV.

[SingleCell]

I have a question about this module. I would like to use it to better calibrate some oscillators, where I am using a buffered mult and precision adder along with a sequencer, to sequence the offset between 2 Oscillators in a West Coast style complex oscillator setup. One sequencer track controls both oscillators, via buffered mult. A second sequencer track adds an offset to the second oscillator only, via a precision adder. The tracking of my oscillators is not quite good enough to do this as well as I'd like.

I believe this module would be perfect for the job except for one issue. I will often send negative voltages from the second sequencer track, to drop the modulating oscillator into the sub audio/LFO range, to get FM squiggles instead of tonal variations. I thought at first I could apply a -3 oct offset to the secondary oscillator via the Caltrans, so that the 0-7V inputs could still reach low enough frequencies (but sadly limiting my upper octaves), but from what I have read, I think it still wouldn't work. It seems it won't be able to calibrate the oscillator down to sub audio frequencies, and will thus send a higher octave that it can accurately track, instead of the LFO speeds I am looking for.

Is this correct? Or is there a way to get the effect I am looking for from the Klavis Caltrans?

[donutschool]

Is this correct? Or is there a way to get the effect I am looking for from the Klavis Caltrans?

I think you're right - it won't drive VCOs outside the calibrated range, and that bottoms out at 20Hz (or -3 V).

Of course - you can always crank down the tuning on the VCO (or use an LFO mode switch if the VCO has one) but touching that dial could invalidate the calibration. If you're *either* using LFO rates or audio rates that might not be a problem - unless the exact frequency of your LFO squiggles is crucial you can just tune back to audio range when you're done - but if you want to smoothly shift between audio and LFO that could be a deal-breaker.

[SingleCell]

Is this correct? Or is there a way to get the effect I am looking for from the Klavis Caltrans?

I think you're right - it won't drive VCOs outside the calibrated range, and that bottoms out at 20Hz (or -3 V).

Of course - you can always crank down the tuning on the VCO (or use an LFO mode switch if the VCO has one) but touching that dial could invalidate the calibration. If you're *either* using LFO rates or audio rates that might not be a problem - unless the exact frequency of your LFO squiggles is crucial you can just tune back to audio range when you're done - but if you want to smoothly shift between audio and LFO that could be a deal-breaker.

Yeah thanks. I think I better learn to embrace the funkiness that is my oscillator tracking, and work within it's limitations.

[ZenPunkHippy]

For this use case I would use some kind of (manual) switch so that the CV for precise tracking via CalTrans is on one route, and the LFO / FM weirdness is selectable from another path. Joranalogue Switch 4 or a matrix switcher would do the job nicely.

[Gringo Starr]

I'm wanting to get this to take in sequences and then to be able to transpose them up in semitones. If I have a sixteen note sequence and transpose it up a 3rd or 6th will it track the transposed sequence perfectly?

[Zentek]

I'm wanting to get this to take in sequences and then to be able to transpose them up in semitones. If I have a sixteen note sequence and transpose it up a 3rd or 6th will it track the transposed sequence perfectly?

it will transpose perfectly if properly calibrated first.
I understand that you will apply the transpose via the CalTrans semitone encoder.

[Morim]

Has anyone tried if you can map the A-196 PLL with CalTrans to use it as a "normal" oscillator?

[Ishkash]

I absolutely LOVE my Caltrans. It is surely the "rug that really ties the room together" - the rug of course being 4 of my favorite oscillators for melodic tones and music.

My one request is surely a silly one overall, but I'll request it anyway: Inverted (black) faceplate! Preferably one that can be separately ordered and installed

But really, it's a wonderful tool and just a joy to work with. Bravo Klavis.

[xidnpnlss]

Has anyone officially tuned a Maths channel with CalTrans? How did that work out for you?

[Bishop Dust]

Has anyone officially tuned a Maths channel with CalTrans? How did that work out for you?

I did it a while ago and it tracked well

[Chupitnik]

Hi, I wonder if you could help me. Three months ago I bought a CalTrans to use with the Twinwaves mk2 for octave changing. I thought the added advantage of oscillator calibration would be good. The problem is that I can't get the device to give accurate results. If I calibrate and oscillator per the instructions all of the notes are either sharp or flat. I tested it by putting 1Volt in from a fixed power source and found that as I changed octaves the output deviated by quite a lot. I tried a neutral calibration and had the same problem. The voltage in is not accurate on output.This is true of all of the outputs and curiously the deviation differs on each one despite being neutrally calibrated and fed a fixed voltage. I have the latest firmware installed. I have also tried the module connected to the power supply on its own and on a second power supply to be sure it's not a power issue.

I made a video showing part of the problem.



After three months of trying I'm beginning to think there may be a problem with the module.

Any help you can give would be greatly appreciated.

Thanks.

I might have double posted this, sorry. I'm very new to all this forum stuff.

[donutschool]

Hi - I've got a Caltrans and a Twin Waves (Mk 1) and, once calibrated, the octave knob will cover 5-6 octaves within about 3 cents.
Sounds like your unit may be faulty (if you have eliminated a power supply problem). I initially tune the oscillator to about C3, not "the lowest note you need to play in tune" or you lose the top end of the range, but I don't think that's your problem.

Sounds like you need to get your unit checked by Klavis

...that said: if you just want an octave switcher for a fairly well-behaved VCO like Twin Waves then I'd stick with the Beast's Chalkboard exp. if you already have it and it does the job. The Caltrans can do a lot more (which makes it look way better value than the chalkboard) but at the expense of complexity and adding extra A-to-D and D-to-A stages to your CV path, which all have their potential distortions. In particular, the input can't accept negative voltages (although it can generate -ve outputs once it is v/oct calibrated).

NB: Public Service Announcement for others: always check the published accuracy of your meter (however many digits it has) - that (reassuringly expensive) Fluke advertises "+/- 0.05% + 1" - so about +/- 4mV on a 5V CV, or about 4 cents over 5 octaves, which should be "good enough for analogue", but even the lower-end Flukes might not cut the mustard (you're paying for a more honest accuracy specification!), let alone your typical Amazon no-name special. If you don't want to put out £600+ for a meter, you're probably better off using a well-behaved VCO and a tuner/frequency meter.

[Chupitnik]

Hi - I've got a Caltrans and a Twin Waves (Mk 1) and, once calibrated, the octave knob will cover 5-6 octaves within about 3 cents.
Sounds like your unit may be faulty (if you have eliminated a power supply problem). I initially tune the oscillator to about C3, not "the lowest note you need to play in tune" or you lose the top end of the range, but I don't think that's your problem.

Sounds like you need to get your unit checked by Klavis

...that said: if you just want an octave switcher for a fairly well-behaved VCO like Twin Waves then I'd stick with the Beast's Chalkboard exp. if you already have it and it does the job. The Caltrans can do a lot more (which makes it look way better value than the chalkboard) but at the expense of complexity and adding extra A-to-D and D-to-A stages to your CV path, which all have their potential distortions. In particular, the input can't accept negative voltages (although it can generate -ve outputs once it is v/oct calibrated).

NB: Public Service Announcement for others: always check the published accuracy of your meter (however many digits it has) - that (reassuringly expensive) Fluke advertises "+/- 0.05% + 1" - so about +/- 4mV on a 5V CV, or about 4 cents over 5 octaves, which should be "good enough for analogue", but even the lower-end Flukes might not cut the mustard (you're paying for a more honest accuracy specification!), let alone your typical Amazon no-name special. If you don't want to put out £600+ for a meter, you're probably better off using a well-behaved VCO and a tuner/frequency meter.

Hi, yes I agree about meter accuracy. In this case the meter is spot on when the Caltrans is not in circuit so the module is really the only factor as far as I can tell.

I have tried so many permutations of calibration procedure but every time the figures are way out. The suspicious thing as that if I do exactly the same calibration procedure on each of the four channels with the same oscillator and voltage source they are all different.

I bought the Beasts Chalkboard and it’s just doing what I want with no fuss. I think it’s time for the Caltrans to clear its desk.

In fairness and for balance I have heard of many happy customers for this product.

[xidnpnlss]

Has anyone officially tuned a Maths channel with CalTrans? How did that work out for you?

I did it a while ago and it tracked well

Nice.

And by any chance did you determine whether the two Maths channels can then be used from the same CalTrans channel? So if I wanted to use the other channel instead of the one for which I calibrated the tuning? Maybe a silly question but I have this idea Maths channels could potentially be different in how they respond to to pitch voltage...