so, i wanna "hack" my vg88.....

[pinkjimiphoton]

got one used for cheap recently and had a go with it.
some things i love about, other i hate... like i hate the fact that no matter where ya play on the string physically, the tone doesn't change.
i like how i can make it feedback.
but i HATE the subharmonic noise crap, so i did some research and discovered a guy named wayne jones had devised a little box to go between the gk pickup and the unit.
but he doesn't make 'em anymore, and i am way too lazy to go thru the hassle of harrassing phil to make me up a board layout

so i downloaded the repair tome and had a looksee, after finding this on the v -guitar forum

RMC wrote>
VG-88 SUBSONIC SENSITIVITY REDUCTION MODIFICATION (RMC - 2001)

-=- FYI, in the VG-88 there are only 6 capacitors to be changed.
The capacitor numbers are C138, C166, C167, C174, C175 & C176. These
are 10uF capacitors which can be changed for 0.22uF caps in order to have a
-3dB subsonic rolloff (single-pole) at about 35Hz (instead of 0.7 Hz !...).
Note that 0.18uF caps will provide -3dB @ 40Hz but they may be a little
harder to find. In this low-frequency filtering application, the film,
ceramic or electrolytic (polar or non-polar) types are all good choices.

now i'm thinking a guitar doesn't have much signal below about 50hz
so changing out the input caps does make sense to me to limit the low frequency noises this things so susceptible to.

so i'm thinking you want to have a range that will cover about an octave drop on the low e, and maybe a 3rd down on the high e.

not being particularly bright or versed in magic myth or science lore,

it seems to me that it would work better if EACH string had its own size of cap.

if i went with say 150n  for the low e, that should cover (ballpark'd, of course) me down to a low enough pitch to improve the tracking, reduce the noise, and still be able to whammy ridiculously.

so by my unschooled logic, the d string, being about an octave higher than the low e should then get a cap around .082n, and then the high  e about 47n, right? being only two octaves apart, this orter work i would think.

so i'm thinking around 100n for the a string, and around 68n for the b string.

does this make sense? the suggestion is to replace 'em all with the same sized cap, but it seems to me that that's not gonna fix the subharmonic problem across the fingerboard.

i figure it should be better to limit the bass frequencies of the higher strings more than the bass strings, as it should be less subharmonic crap to muddle up the system with.

i know its an off the wall kinda question, but...


what do YOU think before i go carving this venerable beast up?

the subharmonics render much of the good stuff about this beast literally unuseable.

thanks guys
respect!!

[anotherjim]

Makes sense to me Jimi. Though as you play higher up the fretboard, subs could come back, though they do seem strongest between 5th fret & nut.
I found the service sheet. Missing info for you is input impedance which make the high pass with those 10uF caps - it is 22k.
So 100nF is 72Hz - close to your bottom E & might still be good if you drop D. 82nF is 88Hz, pretty much perfect for the E but don't forget cap tolerances. My hunch is that 100nF would catch it well enough, but it's just a hunch.
http://sim.okawa-denshi.jp/en/CRhikeisan.htm
Poke values into this calculator. 22k for the R. Write cap value as XX.Xn for nano farads. eg for 4n7 write 4.7n

[pinkjimiphoton]

thanks for the help, jim

i wanna be able to track pitch down to about a b on my low e string so it can track some wiggle stick action.
i find the subharmonics (at least on my les paul, haven't tried my strat yet) seem to be worst thru the middle of the neck.
so i'm hoping that by limiting the bass on  a per-string basis that will help me achieve a more even response, too.
i expect i'll still get "ghost" aliasing subharmonics, but the way it is right now any movement against the body ... even hand movement on strings or overzealous hammerons can trigger bloody horrible ultra low noises.

that's why i was thinking around 150n.... gives me some "wiggle room" for divebombs.

am i correct to think being ballparked around half the capacitance will shift me up around an octave in the audible region?

i don't think i understand the 22k resistance... is that something i should add to the circuit when i change the caps, or is that the load of the circuit having an effect on where the filter cuts off? sorry for all the dumb questions bro.

i'll have a play with the calc you linked when i get back from getting my neck stretched at PT.

think i'm gonna have a crack at this tonite. if it works well, i may explore doing it to my synths as well, as some kinds of patches they seem to suffer from the same affliction.

thanks again jim

[anotherjim]

am i correct to think being ballparked around half the capacitance will shift me up around an octave in the audible region?

It Certainly will.

i don't think i understand the 22k resistance...

Ah, the 22k is the input resistor of an inverting op-amp. That makes the R of the input RC filter (ignoring the input source impedance which I don't know).
Easy to see in the official manual (page 19 of the pdf) I found for d/l here...
https://elektrotanya.com/roland_vg-88_sm.pdf/download.html
Follow the GK-IN connector on that p19.


***BTW, and just to be sure, all of that ^ only applies to the hex pickup - it has no effect on the normal guitar input. For that, I'd suggest reducing the 10uF C81 using 6k resistance in the calculation instead of 22k (actually it's 6k8 in parallel with 100k if you want to work it). 470n there gets you 56Hz.***

[pinkjimiphoton]

thanks bro.
i hadn't even thought about the guitar input, i've only tried it with the hex so far.
that is awesome information, and thanks for explaining opamp input impedance. i'd seriously never even thought about
that aspect before.
i have an order of a wide range of monolythic ceramic caps coming in a couple days, so when they come in i'll commence to hacking on it.
gonna also include the "fx loop" hack on vguitar forum, lets you obtain the straight signal for your guitar from the 13 pin cable, and makes it so the guitar in becomes an fx send instead. you can't use the guitar in and out simultaneously, so it looks like a cool hack.

i will post back when i have news. again jim, from another jim, salud!!

[pinkjimiphoton]

well, the deed is done.

i bricked the fuggin' thing. or so i thought. a quick factory reset and all was well again, i hadn't programmed anything into it yet anyway.

when all was said and done, with the skinniest string being #1 (i hope)
i used the following... figuring they would be close enough for rock n roll.

string 1,(e) c138 47n/.047uF
string 2,(b) c166 56n/.056uF
string 3,(g) c167 68n/.068uF
string 4,(d) c174 100n/.1uF
string 5,(a) c175 150n/.15uF
string 6 (e) c176 220n/.22uF

my immediate thoughts are after messing with it for the last couple hours that its a worthy improvement. i have a feeling i should have gone smaller than i did for the higher pitched strings.
they still have some of that annoying subfrequency noise, so i'd imagine high to low .022, .033, .047, .1 , .15. .22 would be even better.
can't say for sure without testing further, but i am pretty lazy and for my uses i think it's close enough for rock and roll. i bet going right up the scale in decades with the nearest closest value for whatever caps is fine.
it definitely tightened up the tracking of the low strings noticeably.

fwiw, i used all regular ceramic discs for the caps. they work great, are small, and @#$% them 10uF electros

i couldn't find c81 so didn't change that. may be a smd, most of the stuff is other than a few components.

so... another jim from another jim, thanks bro. huge improvement!!!

[anotherjim]

Good going Jimi.
I see there's a lithium coin cell in there for memory backup. You may have accidentally grounded that - the bigger part of the can is the + side so easily done. Scrambled it's memory so it didn't know what to do without a full reset?

Those input caps look like the only place you can easily use to control low freq's. As you make them smaller, you will lose amplitude of the string fundamental - you could go too far with it. But, it looks like it can switch 4 different amounts of gain going into the analog-digital convertors using 4066 switches, so the things "brain" may be able to dynamically boost input somewhat to find best signal.

I found C81. Pdf p16. R/h side of pcb find "Guitar_in" jack then read right to left you have D4, IC18 then C81. Looks like it is a thru hole electro.

[pinkjimiphoton]

hi jim,
yeah, pretty sure thats what i did. i tried to "reboot" it ... sometimes that works with digital stuff that "scans" for any hardware changes.
but a simple reboot did the trick.

i figured too small cap values would do exactly that, but there's gain adjustment for each string individually.
the funny thing is i've found in fuzzes that a smaller amount of bass at the input usually leads to a stronger midrange and bite on the output.

the change did make a huge difference. of course the best way to see would probably use a spare pickup and mount it "backwards" and see if the ghost frequencies move  i guess, but you know me...

pretty lazy!!
thanks for the help with this one bro!!

i think i'm gonna leave the guitar input stock, as i think i'll likely use if for the input of my pedalboard if i ever get to the point of using it live. i don't wanna roll off bass frequencies from the guitar, just the hex pickups which i think has seen a 60-80% improvement overall.

the other thing is some of the traces are wicked small so i don't want to inadvertantly brick it with a bridge somewhere i won't even be able to see.

rock on!!!

[Digital Larry]

i used the following... figuring they would be close enough for rock n roll.

string 1,(e) c138 47n/.047uF
string 2,(b) c166 56n/.056uF
string 3,(g) c167 68n/.068uF
string 4,(d) c174 100n/.1uF
string 5,(a) c175 150n/.15uF
string 6 (e) c176 220n/.22uF

.... but i am pretty lazy ....

Wait, you soldered 6 caps and you call yourself LAZY?  I have to psych myself up for a week to solder anything at all!

[pinkjimiphoton]

i used the following... figuring they would be close enough for rock n roll.

string 1,(e) c138 47n/.047uF
string 2,(b) c166 56n/.056uF
string 3,(g) c167 68n/.068uF
string 4,(d) c174 100n/.1uF
string 5,(a) c175 150n/.15uF
string 6 (e) c176 220n/.22uF

.... but i am pretty lazy ....

Wait, you soldered 6 caps and you call yourself LAZY?  I have to psych myself up for a week to solder anything at all!

lmao.... dude. keep building. 6 caps is nothing

[Digital Larry]

lmao.... dude. keep building. 6 caps is nothing

I've been soldering stuff since I was 7.  Uh... that makes 50 years (wheeze).

This year marks my reluctant return to hardware engineering.  I designed a PCB and so far built the first one by hand.  It is SMT.  Sharpness of eye and steadiness of hand required.  I don't enjoy it quite as much as writing software, and I figured out why.  In software you can really screw things up but it's about the same amount of effort to fix it as it was to do it in the first place because you're just moving text around and then pressing the "go" button.  When you mess up a PCB design sometimes you just gotta throw them in the trash.

Now if I could just get NOS AC128s in SOT-23...