how do I interface to use a JFET as a switch?

Hey all;

I have a guitar effects pedalboard that switches the effects in and out via relays.The switching is set up like this:

Momentary Contact footswitch => 74C373 octal latch => ULN 2003 relay driver => relays and LED indicators

This works really well - does exactly what I want it to, but the relays are a bit on the noisy side (they seem to have gotten worse over time) and click or pop when I switch them. I got the idea that I might be able to replace the relays with a simple transistor switch, and then Eeyore gave me some much better ideas using a JFET. The part number is in the J100 family, and I found the J105 has a 3 ohm on resistance, which should work out nicely. ( A schem of the JFET switch is on ABSE under re:quick emitter follower question). The only problem I'm having now is working out the interfacing of my current system with the JFETs. For starters, Eeyore suggested I use an open collector bipolar to drive the JFET, as opposed to the ULN. Any idea why this might be? Also, the J105 is going to need about -10v to turn the JFET off. Is there a way to interface the positive pulse sent out by the latch to something that will control a negative pulse? I'm picturing something like this:

Momentary Contact footswitch => 74C373 octal latch (positive voltage) =>open collector bipolar (positive or negative voltage?) => J105 (definitely needs negative voltage).

I've never really worked with JFETs so I'm a little in the dark in this area.

Any comments or suggestions are welcome.

Thanks

Reply to
tempus fugit
Loading thread data ...

I don't know what kind of noise/cross talk/distortion this might impose, in fact I didn't even go look at the datasheet, but did you take a look at the CD4066 and its variants? A multi channel analog switch that can be turned on/off using logic level control. Just google the part number and take a look.

Jim

Reply to
James Beck

Thanks for the replies.

When I first started working on this project, I had fully intended to use some type of analog switch. I looked at trygin things with multiplexers, etc., even got some ideas on paper before I realized there was a potentially serious problem. The on resistance of these types of devices is around 80 -

250 ohms. If I had only 1 effect on (and did not switch both input and output) I would only have 80 ohms in the signal path, which may or may not degrade the audio. However, if I had 3 effects on, I would have in the order of a several hundred ohms, which I think would cause at the very least an audible change in the signal leveif not a serious degradation of the tone. If I'm wrong in this assumption please let me know, because both of you guys are right - this would be way easier with some type of analog switch.

Thanks

via

driver

are

click

replace

and I

( A

interfacing of

an

idea

the

=>open

needs

Reply to
tempus fugit

Hi Jamie;

My signal goes from guitar => 1st pedal => 2nd pedal etc => guitar amp, so there is no amp at the start of the chain. Is that what you were picturing when you said there should be no problems?

Thanks

use

potentially

80 -

not

order

an

tone.

guys

message

some

the

controlled

Reply to
tempus fugit

you would gain a lot of saved time by using something like the old stand by 4066 chip which is a bi-directional analog switch controlled by a simple logic level input. if you really want to get fancy with this, you could make it so that it latches/unlatches near the zero crossing point of the current output via comparator! but that would complicate things i know :)

--
Real Programmers Do things like this.
http://webpages.charter.net/jamie_5
Reply to
Jamie

if your driving an amp/pre'amp etc... you should not have any problems. actually the on R would help to prevent cross over impedance problems with other devices. the idea is to do this at the preamp stage and not at the main power stage.

--
Real Programmers Do things like this.
http://webpages.charter.net/jamie_5
Reply to
Jamie

Arp.. switching Audio is a pain in the bootocks. Don't suppose you could fit some nice bright shiny new relays....?

Eeebore probably wanted you to.......

Anyway all of that digital stuff is just an invitation for clicky shit what will detract from how crap your guitar playing is.

What you really want, since you are not bothered about power, is some of those VactecVactrols

formatting link

And... once you have seen one of them you might be tempted to make your own. All you need to do is find a source for light dependent resistors, LDR's. LED's are easy. Then you put them in tubes, glue em up and heatshrink em to keep the light out.

Then you adapt your original relay thing to work with them.

Job Done.

Excellent

DNA

Eeeeh the days of the ORP12.

Reply to
Genome

Bonza

formatting link

DNA

Reply to
Genome

Interesting idea - thanks Genome. Kind of a slow response time though, and even worse turn off (35 ms!!)

Hold the phone - Inspired by previous posts I did some Googling and came up with these from Maxim:

formatting link

formatting link

formatting link

Now I would think that the 2-10 ohms that would be introduced to the signal line should have no effect on the tone. Wouldn't this be the equivalent of having a resistor of the same size in series? would this affect the sound?

Thanks

formatting link

Reply to
tempus fugit

Not my problem... but nice cross fading scenario.

I'm sure you pick your geetar at about thirty picks per second but the rest of us have to spend time finding D having got used to E for 12.

DNA

Reply to
Genome

This is why 9 bar variable Blues was invented. You use the time to draw breath for the next verse whilst getting your fingers right and it gives the drummer time to have a pull on his pint.

It's variable because chances are he'll lose his sticks so you'd have to drop down to 7 but since you are doing 9 variable no-one notices.

DNA

DNA

Reply to
Genome

The switch resistance won't be a big issue.

What *will* be an issue is that an ac coupled signal will go 'below' ground and the bilateral switch won't like that on a single supply. In fact they don't like their ins and out close to supply rail at all. Look for the graph that shows Ron vs input voltage.

A jfet doesn't have this problem but as noted elswewhere a CdS/CdSe photocell might be easier to use. Farnell stock the very good Silonex parts btw.

formatting link
I have no idea why it says 'unbranded' there !
formatting link
For reference that shows what it looks like.

Graham

Reply to
Eeyore

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.