Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
CV recording out of tune
#1
I've been trying to build a generative patch around the Nerdseq.

Initially, I created 3 empty patterns, playing on tracks 3, 4 and 5.
track 1 has each of these patterns on it's own 'island'.
I'm using input one to control the cursor y position, and recording is set so that CV at input 2 will be recorded onto track 1.
The cursor is on track 1, and the idea is that I can use various external to control what is recorded and when onto the three tracks, which are constantly playing back.
I took the cv input from various other modules and tweaked it all until it was harmonious (to me anyway)
The basic concept works ok - like a weird complicated shift register (making the three patterns different lengths, and playback at different clock divisions etc.)

The problem is that the next day, everything seemed off - the y position, the note CV, everything. So I power cycled, and left my system to make sure it was warmed up. Still flakey.
The y position input seemed to be ok sometimes, and sometimes not. It seemed to fix it if I unplugged the patch cable before restarting, but this might just be coincidence. It's still a little variable, but not an issue in this use case.

The input CV tuning is a problem though. In order to try and narrow down the cause, I tried using Nerdseq as the CV source, so I created a three note pattern on track 2, with C1 E1 and G1 cycling (need to be low for the feedback into the inputs to make sense with C-5 being normalised to 0). What I'm getting recorded is mostly C#1, F-1 and G#1, but sometimes I get G-1, and occasionally the others change as well. No consistency.
I've checked that my case power is up to the job, and it *should* be - according to modulargrid specs, the modules in there add up to a few hundred mA less than the Doepfer spec for my power supply...
I haven't touched any calibration controls, And it was a brand new Nerdseq - not a used one, so it's still factory spec.
I also tried changing the trigger delay setting for track 1, but that didn't make any difference.

Any thoughts?
Reply
#2
Depending on the source it makes sense to manually calibrate the inputs so they match the source. Please check the manual for that.
It sounds more to me like that the notes are in between 2 values and probably produce a different result cause of that. I think a calibration of the inputs will fix your issues.
PLEASE use the search function if something have been asked or discussed before.
Every (unnessesary) forum support means less time to develop! But of course, i am here to help!  Smile
Reply
#3
(07-26-2020, 08:05 PM)XORadmin Wrote: Depending on the source it makes sense to manually calibrate the inputs so they match the source. Please check the manual for that.
It sounds more to me like that the notes are in between 2 values and probably produce a different result cause of that. I think a calibration of the inputs will fix your issues.

Thanks for the fast reply.

The notes are coming from the Nerdseq, so I wouldn't expect them to be between 2 notes. Basically, notes from one Nerdseq channel being recorded directly into another another.
I'll have a read through the calibration steps in the manual.

Thanks
Reply
#4
Also the note from the nerdseq can be off if the input is not matched to it.
Just re-calibrate the inputs using C-1 and C-8 from the NerdSEQ and it should solve your problem.
PLEASE use the search function if something have been asked or discussed before.
Every (unnessesary) forum support means less time to develop! But of course, i am here to help!  Smile
Reply
#5
(07-26-2020, 09:35 PM)XORadmin Wrote: Also the note from the nerdseq can be off if the input is not matched to it.
Just re-calibrate the inputs using C-1 and C-8 from the NerdSEQ and it should solve your problem.

So manual calibration?
I tried the auto-calibration a couple of times and it didn't help.
I'll give manual a go and report back.
---
EDIT:
Manually calibrating the inputs to an output seems to have worked. Seems odd that auto-calibration didn't, but as long as something works Smile

Thanks Thomas.

EDIT2: I spoke too soon it seems. The calibration has drifted off now. I'll have to do some more experimenting to see if I can get it stable. Could it be temperature sensitive?
Reply
#6
Today I powered up leaving the calibration as it was yesterday. I had manually tuned it to C-1 and C-8, and it held that for at least half an hour. But then after having left it powered on, but not using it, the tuning/calibration had drifted - maybe over an hour and a half...

Today, immediately after power on, the input was reading A-0 <-12> for C-1 and D-8 <-15> for C-8.

I took reading periodically to chart the changes over time as follows:

Code:
1619:    power on
1619:    A#0 : <-12> and D-8 : <-15>
1625:    A#0 : <-5>  and C-8 : <15>
1637:    A#0 : <8>   and C#8 : <6>
1655:    B-0 : <-15> and C#8 : <-1>
1709:    B-0 : <-8>  and C#8 : <-6>
1738:    B-0 : <1>   and C#8 : <-12>
1800:    B-0 : <4>   and C#8 : <-14>
1844:    B-0 : <12>  and C-8 : <15>
1909:    B-0 : <16>  and C-8 : <11>
1955:    B-0 : <17>  and C-8 : <10>
2037:    B-0 : <15>  and C-8 : <14>

It seems to be settling now after 4.6 hours, but I'll leave it on and keep an eye for changes.
Reply
#7
That seems weird in a way. There should be no drift or at least none that would be significant for either sequencing or recording. If there is no other issue then it sounds to me like a hardware problem with your device. I’m gonna set up a test tomorrow just to be sure (i did these tests before, but i will do it again).
PLEASE use the search function if something have been asked or discussed before.
Every (unnessesary) forum support means less time to develop! But of course, i am here to help!  Smile
Reply
#8
(07-27-2020, 09:09 PM)XORadmin Wrote: That seems weird in a way. There should be no drift or at least none that would be significant for either sequencing or recording. If there is no other issue then it sounds to me like a hardware problem with your device. I’m gonna set up a test tomorrow just to be sure (i did these tests before, but i will do it again).

I'm guessing it's a temperature related issue. My unit is now drifting back in the other direction - And it's got cooler in the room its in.

Ok, Can you suggest any other tests I could do.
I was planning to try moving my Nerdseq to a different case and see if that makes a difference.
Reply
#9
What is the source sequencer/module that you are recording from?
If it is analogue then it might need to warm up. Also, is it clocking the NerdSeq, or is the NerdSeq clocking it?
Reply
#10
(07-29-2020, 07:02 AM)Marizu Wrote: What is the source sequencer/module that you are recording from?
If it is analogue then it might need to warm up. Also, is it clocking the NerdSeq, or is the NerdSeq clocking it?

For testing, I am recording directly from the Nerdseq. No other modules involved. Nerdseq -> Nerdseq.
So setting up a pattern playing C-1 or C-8 (slowly alternating between the two is more practical), patching the output of that channel into an input, and looking at the calibration screen to see what note(s) is being detected. The detected note is changing slowly over time (more quickly at power up).


I have now tested in a different case with no other modules present. In this situation, the calibration settled more quickly - in about half an hour. But way off initially and still slowly changed over time after settling. I suspect will be different on different days depending on ambient temperature. I will power up this case, leave it on for half an hour and calibrate it to C-1 and C-8, then see if it drifts and check again tomorrow to see how far out it is.
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)