Leaderboard

JS Styled High Pass Filter, question.

TonyFlyingSquirrel

TonyFlyingSquirrel

Master Member
Messages
4,519
In attempting to accomplish the wiring on this: http://www.dimarzio.com/sites/default/files/diagrams/JS1200.pdf

I do not see where Stew Mac sells a capacitor with the designated value of .330pf.

I'm wondering, if I could get close by stacking two of their .180pf caps in paralell, and accomplish the same thing.  If so, the value would only be different by a value of .003pf, which I would imagine would be un-noticeable. 

Any more knowledgeable Electronic guru's care to comment?  I must admit, I've always followed designated wire schemes, but haven't experimented with combining multiple capacitors into one switch or pot.
 
I may have misunderstood the "perceived" effect of this wiring. 

For some reason, I was under the impression that it was a bit of a gain cut mod.

It's actually just a switchable treble bleed mod, mostly noticeable at low comes of about 2 or 3.

Anyone know what cap value to do a gain cut?
 
yeah i awas gonna post that the name of that mod is misleading and the mod was a little odd. first if it's designed as a "high pass" that suggests a bass cut. but at 10 volume the cap is shunted so it is a treble bleed. though i think they use a larger cap than the sd treble bleed mod so it would have a bigger effect on low volumes.

now i'm not sure what you really want. gain is something that refers to the ratio of input voltage to output voltage of an amplifier. it's basically the amount of volume. as it applies to guitar amps, it is used to set the amount of over drive, where the gain is set so high that either the signal out of that stage exceeds the supply voltage to that stage, or the output exceed the input range of the next stage so the signal gets clipped making a sine wave resemble a square wave. the nearest thing you can do in a guitar to changinf the gain is changing the volume setting. if you want over drive within the guitar on the otherhand you can install Schottky diodes to ground to clip the signal. if yo uinstall them after the volume, like across the output jack then the volume will act a little like a gain control. but they are more commonly set up on a extra pot or switch so that more overdrive is less signal level... but i don't really know if that's what you are asking...
 
Well, I remember Tony Macalpine had something like this on one of his custom Carvins.

You pull up the pot and it basically was like rolling your volume back well over half way.

Right after I posted last nite I realized the path that the cap in the above schematic was in the same path as the normal treble bleed mod is, just switchable.

So, I wonder if placing it in the output path would be noticeable, or would a resister be more effective?

I recall reading that Neal Schon had some kind of mid-cut switch in his LP Siggy that accomplished the same kind of result, but I can't find a schematic anywhere to determine path and cap/resistor value.
 
so you need some kind of attenuator. to do it with reisistors say you want a 3db drop, which isn't much but it's half the voltage so the math is easy. you might use a larger pot value, say 1 meg instead of 500k then on the output you could have 2 500k resistors in series to ground so they are 1 meg total and in parallel to the volume pot making the load 500k like you would have had with a normal pot selection. you'd have a switch off the output if of the pot and the point the 2 resistors are connected together that selects what point goes to the output.

another way would be to put a resistor between the switch and volume pot that you canshunt with a switch and maybe another resistor to ground to set the load when you use the low setting. there are atually many ways to do this. the first way makes the math easiest but i have a feeling the switch might make a little pop when you switch it.

as far as a mid cut, you either need to add an inductor and capacitor, or a complicated network of resistors and capacitors. you can't do it with just a capacitor. you either need a few of them or an inductor in the circuit. (but maybe one of the pickups could work for that.)

before i start naming values i would need to know how many db you want to drop. 3db is easy because it is very near 1/2 voltage. 10 is easy because it's 1/10th voltage, other numbers are weird. it's based on logrithms which aren't hard but i just haven't done anything with them since highschool. 
 
I'd be willing to go for -10db.

I currently have this in a 500k pot with push/pull switch.  I'm presuming the on/off part of the circuit is located in the same location it currently is, terminals 1 & 2 working from the pot outward as in the diagram above.  But I would have alter where the switch terminates, perhaps right after the output lug?
 
TonyFlyingSquirrel said:
I'd be willing to go for -10db.

I currently have this in a 500k pot with push/pull switch.  I'm presuming the on/off part of the circuit is located in the same location it currently is, terminals 1 & 2 working from the pot outward as in the diagram above.  But I would have alter where the switch terminates, perhaps right after the output lug?
Click to expand...

if you just went from 1-2 like above you'd only be able to make it louder and only at medium volume levels, 10 would always be 10 and the difference between the modes would depend on the position. but there might also be a loss of treble. i'm not really great at photo shop but let me sketch it out latter. i'm not sure if what i'm thinking is exactly the same circuit as featured in the carvin you are talking about but it's one of the simplest ways i can think of attenuating the volume like that.
 
Dan0 said:
TonyFlyingSquirrel said:
I'd be willing to go for -10db.

I currently have this in a 500k pot with push/pull switch.  I'm presuming the on/off part of the circuit is located in the same location it currently is, terminals 1 & 2 working from the pot outward as in the diagram above.  But I would have alter where the switch terminates, perhaps right after the output lug?
Click to expand...

if you just went from 1-2 like above you'd only be able to make it louder and only at medium volume levels, 10 would always be 10 and the difference between the modes would depend on the position. but there might also be a loss of treble. i'm not really great at photo shop but let me sketch it out latter. i'm not sure if what i'm thinking is exactly the same circuit as featured in the carvin you are talking about but it's one of the simplest ways i can think of attenuating the volume like that.
Click to expand...

Thanks so much Dano!
 
schematically it could look like this below, i found this tool on the digikey site to draw that. later i'll try to do it as more of a diagram, it's easier than it looks, a diagram would be less intimidating to look at.

for 3db r1 would be 250k and r2 would be 500k. r3 represents the volume pot at 500k. for 10db R1 would be 450k or something close and r2 would be as close as you can get to 55k. with tollerances in the part and weird schemes for the values i don't think it would pay to get more precise than that in naming values. for 6db you would want r1 around 380k and r2 around 170k. 

 

Attachments

Dan0 said:
schematically it could look like this below, i found this tool on the digikey site to draw that. later i'll try to do it as more of a diagram, it's easier than it looks, a diagram would be less intimidating to look at.

for 3db r1 would be 250k and r2 would be 500k. r3 represents the volume pot at 500k. for 10db R1 would be 450k or something close and r2 would be as close as you can get to 55k. with tollerances in the part and weird schemes for the values i don't think it would pay to get more precise than that in naming values. for 6db you would want r1 around 380k and r2 around 170k.
Click to expand...

So, with a 500k Push/Pull pot, with switch activating from output lug (Lug #2/Center) prior to connecting to the output jack.  It's basically bypassing the normal output path with the attenuated path, sort of like an effects loops does.

From there, I'm unsure of what to put in my shopping cart as far as capacitor/resistor/inductor values. 

I found Fry's Electronics to have a pretty extensive list and found the cap posted in their above link within 3 minutes of walking into their store for the first time.

Thanks again for all your help on this.
 
it's gonna be all resistors. unless you wanted to preserve some extra treble or something. the values aren't nessesarily gonna match the ideal values exactly. 500k might need to be subbed with 470k or 510k or made with a parallel pair at 1meg to get it closer, but the true value might be a 20% tolerance from that.  250k might need to made up from a parallel pair of 510k. you'll have to look through everything and see how you can get close. don't worry about watts ratings or anything. there isn't any real current or power in a guitar do the watts ratings don't matter. it's all overkill.
 
Dan0 said:
it's gonna be all resistors. unless you wanted to preserve some extra treble or something. the values aren't nessesarily gonna match the ideal values exactly. 500k might need to be subbed with 470k or 510k or made with a parallel pair at 1meg to get it closer, but the true value might be a 20% tolerance from that.  250k might need to made up from a parallel pair of 510k. you'll have to look through everything and see how you can get close. don't worry about watts ratings or anything. there isn't any real current or power in a guitar do the watts ratings don't matter. it's all overkill.
Click to expand...

So a 510k resistor with a 1 meg resistor in parallel?  I'm not as picky about the actual values, so long as I can acquire them from Fry's.
 
TonyFlyingSquirrel said:
Dan0 said:
it's gonna be all resistors. unless you wanted to preserve some extra treble or something. the values aren't nessesarily gonna match the ideal values exactly. 500k might need to be subbed with 470k or 510k or made with a parallel pair at 1meg to get it closer, but the true value might be a 20% tolerance from that.  250k might need to made up from a parallel pair of 510k. you'll have to look through everything and see how you can get close. don't worry about watts ratings or anything. there isn't any real current or power in a guitar do the watts ratings don't matter. it's all overkill.
Click to expand...

So a 510k resistor with a 1 meg resistor in parallel?  I'm not as picky about the actual values, so long as I can acquire them from Fry's.
Click to expand...

if you want exactly 500k most suppliers don't make them in that value for some weird reason. to get 500k and not 510k you could use 2 resistors at 1meg each in parallel to each other. but 510k is fine. i wouldn't go to crazy be cause the tollerance is usually larger than that 10kohm. just try to get close.

for resistors or inductors in series the value is the added value. but in parallel you need to take the reciprocal of the sum of the reciprocals. so it's 1/((1/a)+(1/b))
 
Dan0 said:
TonyFlyingSquirrel said:
Dan0 said:
it's gonna be all resistors. unless you wanted to preserve some extra treble or something. the values aren't nessesarily gonna match the ideal values exactly. 500k might need to be subbed with 470k or 510k or made with a parallel pair at 1meg to get it closer, but the true value might be a 20% tolerance from that.  250k might need to made up from a parallel pair of 510k. you'll have to look through everything and see how you can get close. don't worry about watts ratings or anything. there isn't any real current or power in a guitar do the watts ratings don't matter. it's all overkill.
Click to expand...

So a 510k resistor with a 1 meg resistor in parallel?  I'm not as picky about the actual values, so long as I can acquire them from Fry's.
Click to expand...

if you want exactly 500k most suppliers don't make them in that value for some weird reason. to get 500k and not 510k you could use 2 resistors at 1meg each in parallel to each other. but 510k is fine. i wouldn't go to crazy be cause the tollerance is usually larger than that 10kohm. just try to get close.

for resistors or inductors in series the value is the added value. but in parallel you need to take the reciprocal of the sum of the reciprocals. so it's 1/((1/a)+(1/b))
Click to expand...

Would it be better to do it in series, or parallel? 
So, am I going to shop for the following:
1ea) 510k Resistor
2ea) 1 Meg Resistor

Twist the ends together, solder one to the switch post output lug/pre output jack?
 
Dan0 said:
schematically it could look like this below, i found this tool on the digikey site to draw that. later i'll try to do it as more of a diagram, it's easier than it looks, a diagram would be less intimidating to look at.

for 3db r1 would be 250k and r2 would be 500k. r3 represents the volume pot at 500k. for 10db R1 would be 450k or something close and r2 would be as close as you can get to 55k. with tollerances in the part and weird schemes for the values i don't think it would pay to get more precise than that in naming values. for 6db you would want r1 around 380k and r2 around 170k.
Click to expand...

the values are in this post, i was just giving examples about how you may need to work to get there.

you indicated earlier that you might want 10db. so you could use a single 470k to get close to 450k and a single 51k instead of a 55k. otherwise it's a lot of trial and error with the math to get closer and you are within the normal 10% tolerance noted on the resistors anyway. even a 20% tollerance is common. forget the whole parallel resistors, i think you'd actually be able to get close with any of the values i listed above without getting crazy.
 
Dan0 said:
Dan0 said:
schematically it could look like this below, i found this tool on the digikey site to draw that. later i'll try to do it as more of a diagram, it's easier than it looks, a diagram would be less intimidating to look at.

for 3db r1 would be 250k and r2 would be 500k. r3 represents the volume pot at 500k. for 10db R1 would be 450k or something close and r2 would be as close as you can get to 55k. with tollerances in the part and weird schemes for the values i don't think it would pay to get more precise than that in naming values. for 6db you would want r1 around 380k and r2 around 170k.
Click to expand...

the values are in this post, i was just giving examples about how you may need to work to get there.

you indicated earlier that you might want 10db. so you could use a single 470k to get close to 450k and a single 51k instead of a 55k. otherwise it's a lot of trial and error with the math to get closer and you are within the normal 10% tolerance noted on the resistors anyway. even a 20% tollerance is common. forget the whole parallel resistors, i think you'd actually be able to get close with any of the values i listed above without getting crazy.
Click to expand...

I think I got'cha now.
So, output lug to switch, switch (In) to 470k Resistor, 470k resistor to 51k resistor, 51k resistor switch (Out) to output jack positive.

That ought to do it, about a buck & a half at Fry's, and a smack of the mouth & I should be good to go.
 
basically the 51k is going through one side of the push/pull from lug 1 to lug 3 on the volume pot. the 470k is in series with the signal from the pickup selector but gets shunted by the push pull.

i'm not totally sure if you have it or not. technically if the attenuator is on the signal coming from the pickup selector would go to a lug on the push/pull and pass through the 470k where it passes into the 51k and the volume pot simultaneously. it goes to ground through the 51k and to the output jack after the pot. maybe i'm just not following the way you read it back to me. i dunno. i'll do a diagram version latter. both the resistors can go directly on the push/pull. the way i drew it was easy to lay out. maybe i could draw it to better represent how the parts would be assembled with the fewest connections.
 
You must log in or register to reply here.
Back
Top