CV droop with MFOS 1V/oct Keyboard Controller

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: Kent, Joe., analogdigital, infradead, lisa, parasitk, plord, sduck

Post Reply
Matt C.
Learning to Wiggle
Posts: 5
Joined: Thu Dec 31, 2015 10:48 am

CV droop with MFOS 1V/oct Keyboard Controller

Post by Matt C. » Sun Feb 19, 2017 11:11 am

I'm experiencing some pretty serious CV droop issues with my MFOS Single Bus 1V/Oct Keyboard Controller (link to schematic below), I'm wondering if anyone has had similar issues with this board and whether there are any possible fixes.

Right now the pitch CV will drop a half step within probably 4-5 seconds. I tried cleaning the PCB with isopropyl alcohol and a toothbrush but didn't see any obvious improvement. Do I just need to use better cleaner? I'm pretty sure the S&H circuit block is to blame, but if there are other possible causes I'm all ears. Any ideas? Thanks.

Image
You do not have the required permissions to view the files attached to this post.

Praxis
Learning to Wiggle
Posts: 49
Joined: Sun May 29, 2016 9:57 am
Location: UK

Post by Praxis » Sun Feb 19, 2017 12:30 pm

I'm guessing leakage from C19 or C20

User avatar
Nerone
Learning to Wiggle
Posts: 7
Joined: Sat Sep 10, 2016 3:32 am
Location: Rome, Italy

Post by Nerone » Thu May 04, 2017 3:53 pm

Exactly the same problem. Could be an issue in the last stock of PCB???
Help! :cry:

Matt C.
Learning to Wiggle
Posts: 5
Joined: Thu Dec 31, 2015 10:48 am

Post by Matt C. » Thu May 04, 2017 3:56 pm

yeah very frustrating. I've tried everything I can think of, it's gotten slightly better maybe, but the droop is still audible. real bummer.

I'm thinking I might need to install a midi-to-CV converter in order to make this thing actually playable.

User avatar
Nerone
Learning to Wiggle
Posts: 7
Joined: Sat Sep 10, 2016 3:32 am
Location: Rome, Italy

Post by Nerone » Thu May 04, 2017 4:12 pm

I'm using a National LF444, and you?
I'm looking for a test with an IC from another brand (f.i. Texas instruments). Maybe the input resistance is different???

Matt C.
Learning to Wiggle
Posts: 5
Joined: Thu Dec 31, 2015 10:48 am

Post by Matt C. » Thu May 04, 2017 4:15 pm

can't remember what brand my LF444 is. I did try a different chip but it was probably the same brand. Worth a shot to try TI

I had assumed the problem came from a dirty PCB, but after many passes with flux-off, the problem persists.

User avatar
Nerone
Learning to Wiggle
Posts: 7
Joined: Sat Sep 10, 2016 3:32 am
Location: Rome, Italy

Post by Nerone » Thu May 04, 2017 4:57 pm

Me too, tons of flux remover...
So, I'll try it with other components (IC and caps).
Anyway, the analog triggering is fantastic!

User avatar
flagada
Common Wiggler
Posts: 70
Joined: Tue Sep 01, 2015 8:43 am
Location: Amsterdam

Post by flagada » Fri May 05, 2017 2:47 am

What kind of cap did you use for C19?

The analog triggering in mine was actually terrible. If I pressed the same note 10 times very shortly, I could clearly hear pitch differences, sometimes upto half a note. I also had a CV drop, but not within 4-5 seconds, it took longer.

However, I built it on stripboard, so I may have done something wrong. I was glad to switch to a MIDI2CV converter.

User avatar
diablojoy
Super Deluxe Wiggler
Posts: 1615
Joined: Fri Jan 22, 2010 9:03 pm
Location: melbourne

Post by diablojoy » Fri May 05, 2017 4:34 am

What kind of cap did you use for C19?
this !
I have built a few of these boards
you will get some droop eventually but should be after minutes not seconds
for C19 I have always used 10nf polystyrene 0.5% = D tolerance code
obtained from RUSHONLINE

https://www.rush-on-line.co.uk/cap_index.php
this one
LCR SC1276 SK710
It probably wont happen today but if it does it definitely wont go smoothly.

User avatar
Nerone
Learning to Wiggle
Posts: 7
Joined: Sat Sep 10, 2016 3:32 am
Location: Rome, Italy

Post by Nerone » Fri May 05, 2017 5:17 am

I was referring to the triggering speed, in the sense that I feel the difference with the MIDI latency.
In my case it works perfectly, even in fast repeating notes.
The attack pitch is ok, then a very fast droop occurs.
I started with a polystyrene 10n, then I tried with larger caps (not polystyrene), the droop decrease but of course the variations of pitch are not correct, especially in large intervals.

diablojoy, what LF444 did you use? (TI, National ...?)

User avatar
tojpeters
Super Deluxe Wiggler
Posts: 1266
Joined: Mon Feb 06, 2012 10:34 am
Location: Mt.Shasta, Ca

Post by tojpeters » Fri May 05, 2017 8:27 am

I have built 3 of these and they worked great.

JanneI
Wiggling with Experience
Posts: 324
Joined: Mon Oct 19, 2015 3:28 am

Post by JanneI » Fri May 05, 2017 12:44 pm

I've built one which worked great too!

User avatar
diablojoy
Super Deluxe Wiggler
Posts: 1615
Joined: Fri Jan 22, 2010 9:03 pm
Location: melbourne

Post by diablojoy » Fri May 05, 2017 11:19 pm

National LF444CN
I have currently 2 of these boards in use working fine
I'm looking for a test with an IC from another brand (f.i. Texas instruments). Maybe the input resistance is different???
hmm I am thinking this is unlikely to be the culprit

texas instruments aquired national sometime ago so these days should be the exact same chip
data sheet comparison both show 1012input resistance
note this figure is fairly common for JFET input opamps
eg TL074 is the same
LT1058 is perhaps a better option as a replacement for LF444 if you wish to try - more expensive though
also note that polystyrene caps are prone to heat ie keep the leads with enough length to add a heat sink clip component side during soldering and allow time for cooling between doing each leg and be quick.
It probably wont happen today but if it does it definitely wont go smoothly.

User avatar
diablojoy
Super Deluxe Wiggler
Posts: 1615
Joined: Fri Jan 22, 2010 9:03 pm
Location: melbourne

Post by diablojoy » Fri May 05, 2017 11:32 pm

The analog triggering in mine was actually terrible. If I pressed the same note 10 times very shortly, I could clearly hear pitch differences, sometimes upto half a note. I also had a CV drop, but not within 4-5 seconds, it took longer.
That sounds more like the keyboard contacts side of things - KBUS
it is a constant current source through a divider
if your key contacts resistance varies at all you will get pitch differences
what are the contacts made of ?
It probably wont happen today but if it does it definitely wont go smoothly.

User avatar
flagada
Common Wiggler
Posts: 70
Joined: Tue Sep 01, 2015 8:43 am
Location: Amsterdam

Post by flagada » Sat May 06, 2017 7:47 am

diablojoy wrote: That sounds more like the keyboard contacts side of things - KBUS
it is a constant current source through a divider
if your key contacts resistance varies at all you will get pitch differences
what are the contacts made of ?
I used an old Korg Poly-800 keyboard I got from a friend. It had these rubber with carbon contacts. I had cleaned them well with (I think) IPA.

User avatar
diablojoy
Super Deluxe Wiggler
Posts: 1615
Joined: Fri Jan 22, 2010 9:03 pm
Location: melbourne

Post by diablojoy » Sat May 06, 2017 8:55 pm

It had these rubber with carbon contacts. I had cleaned them well with (I think) IPA.


Yep that would explain it then.
that's a known issue with this design - carbon contacts won't work
better off finding an old home organ keybed with metal contacts from the 70's
on ebay quite often you can get them free to good home.
It probably wont happen today but if it does it definitely wont go smoothly.

User avatar
flagada
Common Wiggler
Posts: 70
Joined: Tue Sep 01, 2015 8:43 am
Location: Amsterdam

Post by flagada » Wed May 31, 2017 2:27 pm

@diablojoy: Just saw your reply. Too bad I did not know that before! I spent a lot of time making a nice wooden case for the keyboard and removing a few bad keys. Still managed to make some nice music with it, so it was not totally in vain, though.

Ok, I will keep the module in my modular just in case I find a keyboard with metal contacts.
Image

User avatar
tojpeters
Super Deluxe Wiggler
Posts: 1266
Joined: Mon Feb 06, 2012 10:34 am
Location: Mt.Shasta, Ca

Post by tojpeters » Wed May 31, 2017 4:28 pm

Maybe you can add reed switches and magnets to what you have

User avatar
diablojoy
Super Deluxe Wiggler
Posts: 1615
Joined: Fri Jan 22, 2010 9:03 pm
Location: melbourne

Post by diablojoy » Thu Jun 01, 2017 1:04 am

Maybe you can add reed switches and magnets to what you have
Yes that may also work well
would need to epoxy magnets to underside of keys and figure a way of mounting the reed switches , It's nothing that someone somewhere hasn't done before.
the home organ keybed thing from the 70's is probably going to be a lot easier, though the inverted keys do look nice :tu:
It probably wont happen today but if it does it definitely wont go smoothly.

User avatar
Nerone
Learning to Wiggle
Posts: 7
Joined: Sat Sep 10, 2016 3:32 am
Location: Rome, Italy

Post by Nerone » Sat Jul 08, 2017 9:28 am

FIXED! :banana:
Most likely I cooked TWO pairs of Jfet (Q2, Q3). A carefully soldered new pair and all works fine (just a very slow drooooop......)
Probably the voltage in C19 was sucked back from the -12V line through the gate of the (burned out) Q3 and R14
:doh:

johnstilton
Common Wiggler
Posts: 53
Joined: Tue Oct 09, 2018 10:31 am
Location: West Chicago

Post by johnstilton » Tue Dec 03, 2019 11:57 am

Old thread I know....

I also have this issue on two of these boards that I just finished.

One droops bad. the other droops VERY bad.

I used J310s instead of the recommended 2N5457s. I am going to get some of the latter and some Wima Polyethylene caps and see if either improve the situation.

User avatar
devinw1
Super Deluxe Wiggler
Posts: 1076
Joined: Tue Aug 07, 2018 11:20 am
Location: Portland, OR
Contact:

Post by devinw1 » Tue Dec 03, 2019 12:52 pm

Be very careful when you solder JFETs. Follow good ESD protocol, and for best safety, clip a shunt onto the 3 leads which shorts D, G, and S together while you are soldering and then take this off when you are done soldering.

Post Reply

Return to “Music Tech DIY”