### Author Topic: Why is "ground" reference of boss oc-2 4.9 and not 4.5?  (Read 1125 times)

#### Boner

##### Why is "ground" reference of boss oc-2 4.9 and not 4.5?
« on: March 25, 2019, 03:11:21 PM »
https://www.hobby-hour.com/electronics/s/schematics/boss-oc2-octaver-schematic.gif

I don't understand why the 5.6k/4.7k divider to make "ground" is not smack dab in the middle butthole of 9 volts. Shouldn't it be 4.5?

I'm only guessing, but is to help offset the voltage drop from the germanium diode?

#### MaxPower

##### Re: Why is "ground" reference of boss oc-2 4.9 and not 4.5?
« Reply #1 on: March 25, 2019, 04:33:17 PM »
They probably got a good deal on 5.6k/4.7k resistors. I'd be interested to know why as well. Can't imagine 0.3v matters much considering the tolerance ratings of resistors/caps etc.
What lies behind us and what lies before us are tiny matters, compared to what lies within us - Emerson

#### ElectricDruid

##### Re: Why is "ground" reference of boss oc-2 4.9 and not 4.5?
« Reply #2 on: March 25, 2019, 05:51:39 PM »
My guess would be that the various loads on the midpoint voltage tend to drag it down, so they added a bit of compensation into the resistor values so that it finishes up in the right place when loaded. Have you measured the midpoint voltage in an OC-2? If we did, we'll soon know...

Tom

#### R.G.

##### Re: Why is "ground" reference of boss oc-2 4.9 and not 4.5?
« Reply #3 on: March 25, 2019, 07:27:27 PM »
Could be any of the above. If I had to guess, it would be that the circuit works by turning two JFETs on and off, and they wanted the extra voltage to ensure a solid "off" on Vgs.

Putting a reference voltage in exactly the middle presumes that what you're after is maximum undistorted signal swing. That presumes that your circuits have symmetrical swings (some don't) and that biggest signal swings are the most important thing. But if some other criteria is more important, you can easily move your "AC ground" reference to be not in the middle.

Like so many things in life, it's a question of objectives - what is it that you're most wanting to accomplish, and what will you trade off to get it?
R.G.

In response to the questions in the forum - PCB Layout for Musical Effects is available from The Book Patch. Search "PCB Layout" and it ought to appear.

#### amz-fx

##### Re: Why is "ground" reference of boss oc-2 4.9 and not 4.5?
« Reply #4 on: March 26, 2019, 08:07:13 AM »
Schematic with IC numbers:

The BA634 is an unusual choice. Probably selected for small footprint.

« Last Edit: March 26, 2019, 08:14:49 AM by amz-fx »

#### EBK

##### Re: Why is "ground" reference of boss oc-2 4.9 and not 4.5?
« Reply #5 on: March 26, 2019, 11:52:59 AM »
The Stompboxology article on octave down circuits discusses the voltage on the diodes and explains a bit about why this voltage isn't half-V+.
http://moosapotamus.net/files/stompboxology-freq-div-sub-synth.pdf (info scattered within pp. 12-14).
• SUPPORTER

#### MaxPower

##### Re: Why is "ground" reference of boss oc-2 4.9 and not 4.5?
« Reply #6 on: March 26, 2019, 01:38:11 PM »
Well, there it is. Something to catalog for future reference.
What lies behind us and what lies before us are tiny matters, compared to what lies within us - Emerson

#### Boner

##### Re: Why is "ground" reference of boss oc-2 4.9 and not 4.5?
« Reply #7 on: March 27, 2019, 05:56:03 PM »
The Stompboxology article on octave down circuits discusses the voltage on the diodes and explains a bit about why this voltage isn't half-V+.
http://moosapotamus.net/files/stompboxology-freq-div-sub-synth.pdf (info scattered within pp. 12-14).

Thank you so much!!

So I spent about an entire day trying to get a working simulation of an OC-2 in multisim, but no matter what it would not work. The problem was the cap/diode "ac restorer" part of the circuit. It would work as normal when the circuit ended with just the cap/diode, but when attached to the proceeding opamp/jfet inverter dealy, the DC offset would disappear and the AC signal would be at 0 DC, and not some DC value to keep everything positive or negative.

The only way I could get it to work was to use the op amp version (active and not passive) of the clamper circuit I found on wikipedia.

I havent gone passed the simulation stages, but it does seem to work with vref being 4.5 and has the benefit of having NO offset to take into account when figuring out ground. So instead of using a germanium diode, you can use silicon. I went with 1n914.

https://en.wikipedia.org/wiki/Clamper_(electronics)#Op-amp_circuit

I've attached what I have so people can chime in and/or use it, lord knows I need to contribute something other than sucking out knowledge.