AMZ FX ANALOG ALCHEMY GEOFX MEF ROG MEF DIY WIKI

will this work for diode switching?

Started by jimbob, July 06, 2006, 12:54:24 PM

Previous topic - Next topic
Print
Go Down Pages1
User actions

jimbob

July 06, 2006, 12:54:24 PM
I found this online a while back and wondered if instead of switching between 2 different caps would it also or instead switch between 2 diffferent pais of diodes. Im thinking so but wanted to double check.

Where it says "original cap location I assume one lead would go just before where the diode pait began on the board and the other lead would go right after the last? Im wanting to switch between stock 1n914's and led's for a Rat clone.

Any suggections. Im really just learning about switching and appreciate any help. Oh- please- no links to Geo :) Im there reading too.

Thanks
"I think somebody should come up with a way to breed a very large shrimp. That way, you could ride him, then after you camped at night, you could eat him. How about it, science?"

Peter Snowberg

#1
July 06, 2006, 01:08:54 PM
Correct. 8)

Diodes give you an additional option because the forward voltage is different for different types. The forward voltage of an LED is much higher than a plain silicon signal diode so if you leave the LEDs connected at all times and then switch the signal diodes in, in parallel with the LEDs, the signal diodes will take over. No need to disconnect the LEDs. That way you only need a SPST switch to get the same result.


    +---switch---1N914_pair---+
    |                         |
----+---------LED_pair--------+----

<--      to rest of circuit      -->
Eschew paradigm obfuscation

jimbob

#2
July 06, 2006, 01:33:24 PM
Thanks- Ill give it a try.
"I think somebody should come up with a way to breed a very large shrimp. That way, you could ride him, then after you camped at night, you could eat him. How about it, science?"

Mark Hammer

#3
July 06, 2006, 02:03:33 PM
It'll work.  Keep in mind that diodes are much like any component in the sense that shunting (bypassing them with a straight-wire link) makes them ineffective.  So, f'rinstance, one could have a pair of silicon back-to-back diodes, followed by a second pair of germanium back to back diodes.  If a SPDT switch was situated so that the "common" was connected to where those two pair joined each other, and each outside lug went to one end of each diode pair, pushing that toggle to one side or the other would insert a shunt in parallel with the one pair or the other, you would essentially make one diode pair "active" at a time.

So what is the advantage of this over other switching schemes?  Simple.  Let's say you have a 3-position, center-off, SPDT toggle.  In the middle position, there is no shunt of either diode pair, so its like you have the one set in series with the other.  In one side position, one pair is shunted, and in the other side position the other pair is shunted.  So this switch would give you Si clipping, Ge clipping, or GE+SI in series.

A common way of switching diodes in those pedals where they go to ground (e.g., Distortion+, Bosstone, Superfuzz, etc.) is to connect all diodes used to the same point on the audio path, and use the toggle to make or break the connection between some set of diodes and ground.  So, imagine that you had a pair of 1N914s in place of the blue cap in your left-hand picture, and a pair of LEDs instead of the green cap in the same picture.  The centre lug of the toggle goes to ground.  This would let you select between the clipping action of the one type of diode or the other.  If the switch was a centre-off 3-position type, that would provide a 3rd option of nothing connected to ground.

jimbob

#4
July 06, 2006, 02:17:32 PM
QuoteA common way of switching diodes in those pedals where they go to ground (e.g., Distortion+, Bosstone, Superfuzz, etc.) is to connect all diodes used to the same point on the audio path, and use the toggle to make or break the connection between some set of diodes and ground.  So, imagine that you had a pair of 1N914s in place of the blue cap in your left-hand picture, and a pair of LEDs instead of the green cap in the same picture.  The centre lug of the toggle goes to ground.  This would let you select between the clipping action of the one type of diode or the other.  If the switch was a centre-off 3-position type, that would provide a 3rd option of nothing connected to ground.

I like that idea.

Im sure i did something wrong here because Im not getting any change of sound other than that of what the effects sounds like without ANY diodes. Very loud and a little feedback. What I did was connect a pair of leds to the top 2 lugs in series (one lead went to the left lug of the dpdt and the other end of the connected pair to the other top lug ( right) I did this with the 1n914's to the bottom 2 lugs the same way (such as I understood the pic on the right above. The middle 2 lugs I had 1 connected to ground and the other to a place between the 1k and d2
"I think somebody should come up with a way to breed a very large shrimp. That way, you could ride him, then after you camped at night, you could eat him. How about it, science?"

jimbob

#5
July 06, 2006, 02:24:02 PM
OK- I feel dumb- I figured it out. This time Im using a spdt. Makes more sense now.
"I think somebody should come up with a way to breed a very large shrimp. That way, you could ride him, then after you camped at night, you could eat him. How about it, science?"

Peter Snowberg

#6
July 06, 2006, 02:29:14 PM
If you use a SPDT on-off-on switch, you can get three very different choices very easily with another variation on Mark's suggestion. If you follow the diagram you posted above and replace the green cap with a pair of Ge diodes and the blue cap with a pair of Si diodes, you can wire up a pair of LEDs in parallel and get a Ge-LED-Si mode switch. Using a pair of Ge diodes for each direction will make the output more similar to the Si level. The LEDs will always result in the largest amount of signal getting through because they have the highest forward voltage.
Eschew paradigm obfuscation

jimbob

#7
July 06, 2006, 03:12:54 PM
I wired it up w a spdt w 1 outer lug going to a pair pf 1n914;s and the other lug connected to a pair of led's. Both ends of both pars are connected togather and soldered to where d1/2 are connected to the 1k and the middle lug of the spdt is soldered to the other end on the board where the diode pair connected to ground.

What am I doing wrong here? Of the 3 positions I get an obviosly lower but more distortion. I assume thats the 1n914's but the other 2 positions sound the same ( almos what it sounded like without any diodes.
"I think somebody should come up with a way to breed a very large shrimp. That way, you could ride him, then after you camped at night, you could eat him. How about it, science?"

jimbob

#8
July 06, 2006, 03:38:09 PM
OK- so far I thought I might have had a bad led so I replaced them w some germ diodes and now Im getting closer. Now I have 3 different sounds - the out 2 are less loud but have a nice distortion the middle I assume has nothing there as far as diodes so Its REALLY loud in comparison. Im still trying to figure out how- based on what Ive already done- do you add another to the mix that being led's that actually work.
"I think somebody should come up with a way to breed a very large shrimp. That way, you could ride him, then after you camped at night, you could eat him. How about it, science?"

jimbob

#9
July 06, 2006, 04:01:30 PM Last Edit: July 06, 2006, 04:30:42 PM by jimbob
QuoteSo, imagine that you had a pair of 1N914s in place of the blue cap in your left-hand picture, and a pair of LEDs instead of the green cap in the same picture.  The centre lug of the toggle goes to ground.  This would let you select between the clipping action of the one type of diode or the other.  If the switch was a centre-off 3-position type, that would provide a 3rd option of nothing connected to ground.

Thats how I did it and the centre seems to also sound VERY similar to the option next to it. The other side of the switch is considerably different. Not as loud and more distortion. Is that right?

Quotereplace the green cap with a pair of Ge diodes and the blue cap with a pair of Si diodes, you can wire up a pair of LEDs in parallel and get a Ge-LED-Si mode switch.

" pair of LEDs in parallel" Im not sure I understand?
"I think somebody should come up with a way to breed a very large shrimp. That way, you could ride him, then after you camped at night, you could eat him. How about it, science?"

jimbob

#10
July 06, 2006, 04:32:21 PM
Would all this have worked easier w a 5p2t in which I would just use 3 or 4 of the five ?
"I think somebody should come up with a way to breed a very large shrimp. That way, you could ride him, then after you camped at night, you could eat him. How about it, science?"

Peter Snowberg

#11
July 06, 2006, 04:52:25 PM
Quote from: jimbob on July 06, 2006, 04:01:30 PM
Thats how I did it and the centre seems to also sound VERY similar to the option next to it. The other side of the switch is considerably different. Not as loud and more distortion. Is that right?

Quotereplace the green cap with a pair of Ge diodes and the blue cap with a pair of Si diodes, you can wire up a pair of LEDs in parallel and get a Ge-LED-Si mode switch.

" pair of LEDs in parallel" Im not sure I understand?

If you don't get much change in the sound by removing the diodes, that shows that the opamp is clipping and producing the distortion. ;)

By "pair of LEDs in parallel" I mean two LEDs, in parallel with each other, and with opposite polarity, which are then in parallel with the switch that allows selection of two additional pairs of diodes, and those diode pairs. If no additional diode pairs are selected, you get the sound from the LEDs. If you select either of the other diode pairs, the lower VF of the new pair will cause that new pair to take over the clipping role.

----+---LED_clipper---------------+----
    |                             |
    |                             |
    |     Three-------Ge_diodes---+   
    +----Position                 |
          Switch------Si_diodes---+
   
Position 1: Ge clipping
Position 2: LED clipping (with some opamp clipping too it sounds like)
Position 3: Si clipping
Eschew paradigm obfuscation
Print
Go Up Pages1
User actions