Scaled voltage reference with octave and semitone steps

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

Moderators: Joe., lisa, luketeaford, Kent

Post Reply
User avatar
Haralds:Werk
Wiggling with Experience
Posts: 268
Joined: Fri Sep 09, 2016 3:33 am
Location: Germany
Contact:

Scaled voltage reference with octave and semitone steps

Post by Haralds:Werk » Wed Mar 25, 2020 11:47 am

This module provides high precision CV outputs in 1V (octaves) and 83,3mV (halves) steps. The 1V output goes from -5 to +5V. The 83,3mV steps goes from -5 to plus 5 steps (halves). This module is thought for all who are missing octave switches in some modules. Especially in VCO. With this module you can switch octaves and halves as well.


Specs and features
• High precision output from -5 to +5V in 1V steps (octaves)
• High precision output in 83,3mV steps, +/- 5 steps (Halves)
• Runs on +/-15V and +/-12V
• Power consumption around 30mA each rail


The documentation and the Gerber files for download can be found in my website.
Scaled_voltage_reference_301_schematic.png
The precision voltage is derived from the REF102. The negative voltage is provided with the INA105. It is crucial to match the resistors in the voltage dividers as good as you can. The outputs of the voltage dividers are buffered to avoid loading of the dividers. The resistors around the OpAmps must be matched as well. The one volt and the 83,3mV steps are added together with IC4A. The three outputs are individual buffered.
Scaled_Volt_Ref_side.JPG
Scaled_Volt_Ref_front.JPG
You do not have the required permissions to view the files attached to this post.
Last edited by Haralds:Werk on Sat Mar 28, 2020 3:37 pm, edited 1 time in total.
www.haraldswerk.de
Home of Next Generation Formant

User avatar
neil.johnson
Super Deluxe Wiggler
Posts: 1301
Joined: Thu Apr 08, 2010 3:51 pm
Location: Cambridge, UK
Contact:

Re: Scaled voltage reference with octave and semitone steps

Post by neil.johnson » Wed Mar 25, 2020 2:10 pm

Interesting module. I'm curious why you don't use the REF02 (and it's inverter) to bias the semitone switch as well? You just need to replace the 100k resistors with 40k and everything else stays the same (or replace the 40K+15k with 54k9 resistors, plus trimmers if you really care).

You might also want to consider adding small (100pF) capacitors around the output drivers to tame their response when the load changes or you change the settings. For example, your output configuration is what I call intr in the following post:
viewtopic.php?f=16&t=81658&p=1129644&hi ... c#p1129644

Neil
Modular stuff :: http://www.njohnson.co.uk

User avatar
BugBrand
Knowledge of Bugs
Posts: 7187
Joined: Tue Jul 15, 2008 11:59 am

Re: Scaled voltage reference with octave and semitone steps

Post by BugBrand » Wed Mar 25, 2020 4:11 pm

Really good to see - I made something some years back but was never quite sure of its accuracy. It would be good to DIY something new for my work bench.
Does it really not require any trimmers? Maybe I got carried away worrying about trimmers & offsets.

User avatar
EATyourGUITAR
has no life
Posts: 4038
Joined: Tue Aug 31, 2010 12:24 am
Location: Providence, RI, USA

Re: Scaled voltage reference with octave and semitone steps

Post by EATyourGUITAR » Wed Mar 25, 2020 4:15 pm

I used SIP resistor arrays 0.1% and AD648 but the parts are in a bag unfinished. never had the time or the need for making traditional music. I thought it would be good to have the resistors on a single package so the temp is the same.

edit:

old mouser cart showed these
here https://www.mouser.com/ProductDetail/584-REF01CPZ

https://www.mouser.com/ProductDetail/584-OP177GPZ

https://www.mouser.com/ProductDetail/60 ... 5BRD52-10K

maybe I was drunk the first time. then I did it all over again with the AD648 + resistor array. this is why I don't drink.
WWW.EATYOURGUITAR.COM <---- MY DIY STUFF

User avatar
Haralds:Werk
Wiggling with Experience
Posts: 268
Joined: Fri Sep 09, 2016 3:33 am
Location: Germany
Contact:

Re: Scaled voltage reference with octave and semitone steps

Post by Haralds:Werk » Sat Mar 28, 2020 11:41 am

BugBrand wrote:
Wed Mar 25, 2020 4:11 pm
Really good to see - I made something some years back but was never quite sure of its accuracy. It would be good to DIY something new for my work bench.
Does it really not require any trimmers? Maybe I got carried away worrying about trimmers & offsets.
No trimmers needed. But you have to select the resistors very carefully and use very low offset OpAmps.
www.haraldswerk.de
Home of Next Generation Formant

User avatar
Haralds:Werk
Wiggling with Experience
Posts: 268
Joined: Fri Sep 09, 2016 3:33 am
Location: Germany
Contact:

Re: Scaled voltage reference with octave and semitone steps

Post by Haralds:Werk » Sat Mar 28, 2020 4:04 pm

neil.johnson wrote:
Wed Mar 25, 2020 2:10 pm
Interesting module. I'm curious why you don't use the REF02 (and it's inverter) to bias the semitone switch as well? You just need to replace the 100k resistors with 40k and everything else stays the same (or replace the 40K+15k with 54k9 resistors, plus trimmers if you really care).
Yes, you can do it this way. If you find a 40k resistor. I wanted to avoid trimmers.

neil.johnson wrote:
Wed Mar 25, 2020 2:10 pm
You might also want to consider adding small (100pF) capacitors around the output drivers to tame their response when the load changes or you change the settings. For example, your output configuration is what I call intr in the following post:
viewtopic.php?f=16&t=81658&p=1129644&hi ... c#p1129644

Neil
I use your suggested scheme in buffered multiples or active outputs in modules with rapidly changing (audio) signals. This is not necessary here. The output is a constant voltage and is usually not changed while you play. The load is constant as well. And the maximum voltage change is only 1V per step and not the 10Vpp in your simulation. And the used LT1014 has only a slew rate of 0,4 V/uS. Of course it will do no harm to add the capacitor.
www.haraldswerk.de
Home of Next Generation Formant

User avatar
Graham Hinton
Super Deluxe Wiggler
Posts: 3054
Joined: Fri Jul 02, 2010 6:28 pm
Location: England
Contact:

Re: Scaled voltage reference with octave and semitone steps

Post by Graham Hinton » Sun Mar 29, 2020 9:43 am

neil.johnson wrote:
Wed Mar 25, 2020 2:10 pm
Interesting module. I'm curious why you don't use the REF02 (and it's inverter) to bias the semitone switch as well? You just need to replace the 100k resistors with 40k and everything else stays the same (or replace the 40K+15k with 54k9 resistors, plus trimmers if you really care).
It's easier than that: use the same resistor divider chain and make R29 120k.

Bugbrand wrote:Does it really not require any trimmers?
It's a lot easier with a few in key places. I'm making PitchMaster and hand matching 0.1% resistors. The issue is can you use the ones left over? Do you want an absolute match or a relative one? What I find is that even with 0.02% accuracy a trimmer still helps.

User avatar
Haralds:Werk
Wiggling with Experience
Posts: 268
Joined: Fri Sep 09, 2016 3:33 am
Location: Germany
Contact:

Re: Scaled voltage reference with octave and semitone steps

Post by Haralds:Werk » Sun Mar 29, 2020 11:00 am

Graham Hinton wrote:
Sun Mar 29, 2020 9:43 am
neil.johnson wrote:
Wed Mar 25, 2020 2:10 pm
Interesting module. I'm curious why you don't use the REF02 (and it's inverter) to bias the semitone switch as well? You just need to replace the 100k resistors with 40k and everything else stays the same (or replace the 40K+15k with 54k9 resistors, plus trimmers if you really care).
It's easier than that: use the same resistor divider chain and make R29 120k.
Wow!!!! Excellent.
www.haraldswerk.de
Home of Next Generation Formant

User avatar
Graham Hinton
Super Deluxe Wiggler
Posts: 3054
Joined: Fri Jul 02, 2010 6:28 pm
Location: England
Contact:

Re: Scaled voltage reference with octave and semitone steps

Post by Graham Hinton » Sun Mar 29, 2020 5:02 pm

Haralds:Werk wrote:
Sun Mar 29, 2020 11:00 am
Wow!!!! Excellent.
Sorry to tear half your circuit up.

Also remember that even with 0.1% resistors the error of an inverter can be 0.2%, i.e. one resistor +0.1% and the other -0.1%. So why have a precision inverter on the reference and then go through two 0.2% stages? You can lose IC4A by reversing the order of the rotary switches. The REF02 has at least 0.3% error so that needs a trimmer to get 5.000V, then you can lose the INA105 and use the saved op amp instead.

So that's got the cost down and improved the accuracy.

User avatar
Haralds:Werk
Wiggling with Experience
Posts: 268
Joined: Fri Sep 09, 2016 3:33 am
Location: Germany
Contact:

Re: Scaled voltage reference with octave and semitone steps

Post by Haralds:Werk » Mon Mar 30, 2020 1:33 am

Graham Hinton wrote:
Sun Mar 29, 2020 5:02 pm
Haralds:Werk wrote:
Sun Mar 29, 2020 11:00 am
Wow!!!! Excellent.
Sorry to tear half your circuit up.

Also remember that even with 0.1% resistors the error of an inverter can be 0.2%, i.e. one resistor +0.1% and the other -0.1%. So why have a precision inverter on the reference and then go through two 0.2% stages? You can lose IC4A by reversing the order of the rotary switches. The REF02 has at least 0.3% error so that needs a trimmer to get 5.000V, then you can lose the INA105 and use the saved op amp instead.

So that's got the cost down and improved the accuracy.
Thanks. Got it. Will do if I ever build a next version.
www.haraldswerk.de
Home of Next Generation Formant

Post Reply

Return to “Music Tech DIY”