Enclosure for Remote Switcher/ Mounting DB25?

[tmichutka]

Hey everyone,

  So I'm going to attempt to make the RG remote switcher (http://www.geofex.com/article_folders/rmtswtch/rmtsw.htm) and although I'm kinda starting to feel like I'm in over my head, that's another story, my most immediate concern right now is the housing for the main unit (with all the jacks, relays, etc).  The only Hammond enclosure I could find that might work is their rackmount series, but those are relatively expensive (~$50) and unnecessary since I'm going to be securing this to my pedalboard.  I'm considering just gutting out a graphic EQ I found at Goodwill which has quite a bit of blank space on the back for drilling.  That'll cut down the price pretty significantly.  Any recommendations/comments?

  Another question: I'm going to be using DB25 jacks but I'm not sure how to go about drilling a rectangular hole for them to go in.  I'm guessing it involves something more/other than drilling, and I'll probably sound like an idiot even wording the question like that.

Thanks much!

[R.G.]

You go guy!!

Here's a tip that might help: http://www.industrialmetalsales.com/Aluminum_Rectangle_Tube.html

Industrial Metal Sales will sell you a three foot long stick of 1" by 3"  aluminum retangular tubing for $18.35. Three feet of 2" x 4" is $21.18. There are other sizes and shapes as well. There's some work to be done on it with hacksaw and files, but you wind up with a really pro looking result if you're careful.

As to drilling DB25 holes; there are really only two ways, punch with a special punch and chain drilling. I'm eliminating punching, as no one has the punches, and that leaves chain drilling. To do this, you carefully scribe the outline of the hole you want on the surface, and then drill out as much of it as possible with normal drill bits. You then use a file and cut the rest away to the scribe lines.

Here are some tips:
1. Use masking tape over the material where it's clamped into vises or will be scraped by tools or cuttings. This keeps the surface looking nice and it's easier to clean up if you paint.
2. Before you file, especially if you are filing aluminum, have a fine wirebrush and hard blackboard chalk handy. Clean the file's grooves out with the wire brush and fill as much of the file's teeth with chalk as you can. Aluminum is the world's worst metal for leaving little bits in the teeth of the file (machinists call this "pinning"), which will make filing even harder and leave ugly marks. Clean and re-chalk as you go.
3. Choose between big drill bits that eat all the space you can and small drill bits close together around the edges of the cutout. Use a drill press for the many-holes version, as you need to be sure you don't drill outside the lines.

[tmichutka]

I'm so sorry it took me so long to reply RG!  I'm actually really starstruck that RG replied to my post     I couldn't have asked for anyone better to!

That's awesome, thanks so much for that tip about the aluminum!  Their prices look right up my alley so I'll definitely be using them!  Likewise the DB25 drilling tips... sounds easy enough with the chain drilling.

I couldn't possibly thank you enough for your article on the remote bypass system!  I hope you don't mind if I pick your brain quickly about the schematic... I'm trying to incorporate the pop supressing circuit but I'm having a hard time fitting it physically onto a PCB I'm designing in DIYLayout so I'm thinking about just using the circuit for the LED indicator.  I'm using 5VDC DPDT Omron latching relays (G5AK-237P), so I'm thinking (hoping) the pop will be negligible, or at least lower than the guitar signal level.  Since I'm not using the pop supressing circuit at the +on/-off, would I essentially ignore everything except LED1, D2, and R3 going to the +5VDC, and feed D2 and the LED ground, or is it more complicated than that?  I'm thinking it's not that simple as I write this, but if it's not too complicated maybe you could explain how I could wire this with just LED indicators.  I think once I get past that hurdle I'll be well on my way to getting this built!

I'm really really excited!  This is so overdue because my pedalboard has way outrgrown the tiny stages I play on 

[R.G.]

I'm trying to incorporate the pop supressing circuit but I'm having a hard time fitting it physically onto a PCB I'm designing in DIYLayout so I'm thinking about just using the circuit for the LED indicator.  I'm using 5VDC DPDT Omron latching relays (G5AK-237P), so I'm thinking (hoping) the pop will be negligible, or at least lower than the guitar signal level.  Since I'm not using the pop supressing circuit at the +on/-off, would I essentially ignore everything except LED1, D2, and R3 going to the +5VDC, and feed D2 and the LED ground, or is it more complicated than that?  I'm thinking it's not that simple as I write this, but if it's not too complicated maybe you could explain how I could wire this with just LED indicators.  I think once I get past that hurdle I'll be well on my way to getting this built!

Hmmm...

Sometimes a relay is quiet enough if you just put a capacitor across the coil. It's harder on the driver transistor that way, as it has to eat the full capacitor current, but it does slwo down the edge of the coil voltage. Some relays are quiet enough by themselves, some need a lot of quieting.

I noticed you mentioned the Omron latching relays. Latching relays generally don't UNlatch unless you do something to them, and there's no unlatching built into that logic. Are your relays really self latching? If they are, that driver circuit won't work.

If they're not, I still recommend building up one relay's worth of the circuit on a breadboard or perf and making sure that your driver and relay work well together before committing to a PCB. Iv'e been bitten by that a couple of times myself.

[tmichutka]

Ohhhh, ummm, yeah, they are latching relays... I kinda forgot about the whole latching/non-latching thing.  Mmmm, okay, hmmm.  I'm gonna have to rethink this whole thing.  I guess that's what I get for buying the cheapest thing I could find without checking to make sure it's what I need  .  So, the latching ones are the ones that do not have a normal position... I was thinking it was just the opposite... and I had even read your article about Relays for Switching Audio Signals!!!  That's embarassing.

Is there a page by chance that goes over pop supression for latching relays?  I'm pretty much back at square one now.  Would you recommend just getting non-latching relays or do you think I can make it work with the latching ones?

Again, thank you sooooo much for your help!  I can't tell you how much I appreciate it!

[tmichutka]

Okay so I've ordered the aliminum from industrialmetalsales and I actually found a really good deal on some Coto 12VDC Shielded reed relays on eBay so I'm probably going to get those since you recommend Reed Relays... plus they're shielded!

Oh, one quick question though, do you think it'd be easier just using perfboard, or is a PCB the way to go.  Honestly this is going to be my first PCB design, but it seems like a pretty good place to start since all the work's pretty much done for me thanks to RG & geofex!    I promise that's my last question!

Oh, by the way, if anyone needs 10 latching dual-coil 5VDC relays (Omron G5AK-237P) I'll give you a great deal!      I have no use for them anymore so I'll let go of them for less than a dollar each plus shipping... if you want all 10 I'll let them go for $5 plus shipping!  As you can see (http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&ssPageName=STRK:MEBX:IT&item=350040501622&_trksid=p3984.cALLBUY.m311.lVI it's half what I paid for them a few days ago, but meh, at least someone will get some use out of them.

[R.G.]

Is there a page by chance that goes over pop supression for latching relays?

My "Relays for Audio" has as much as anyplace.

It's pretty simple: there is capacitance between the coil and the audio contacts. The audio contacts have impedances to ground connected to them. The relay coils see full-voltage square waves impressed on them, so that waveform through the capacitance to the signal path impedance causes a "tick" voltage to appear in the audio chain, over and above whatever pops the switching contacts themselves do.

To suppress this, you can go two ways, as suggested by the Ohm's-law nature of the voltage divider of the parasitic capacitance to the signal path. These are (a) make the capacitance smaller, which amounts to making the source impedance higher, and coupling less of the voltage in; an alternate is to make the audio path impedance to ground lower, eating more of the 'tick', but that's usually not possible. (b) slowing down the leading and trailing edge of the coil waveform; this has the effect of lowering the frequency content of the coil waveform, and making the 'tick' content too small to couple well through the small capacitance. You can do (b) by slowing down the driver, as I show with a transistor and a feedback capacitor or by putting a big capacitor across the coil.

That's kind of all there is.

I actually found a really good deal on some Coto 12VDC Shielded reed relays on eBay so I'm probably going to get those since you recommend Reed Relays... plus they're shielded!

And that is a really good deal. Shielded reed relays was the Telephone company's solution, and it's a good one. The shielding makes the coupling capacitance really, really small, so you should be in good shape. Be sure and ground the shields!

Oh, one quick question though, do you think it'd be easier just using perfboard, or is a PCB the way to go.  Honestly this is going to be my first PCB design, but it seems like a pretty good place to start

I think you might do it a step at a time on perfboard. That's a really aggressive PCB setup to try the first time. It might go fine, but misunderstanding some little thing might also make it into a long process of fixing. It is also possible that there is some latent flaw in that layout. It's been a long time since I did that. I'm just being extra conservative, but if you have to have the first and only product work, go with hand work, not PCB work.

I would build ONE channel of jacks/relay, putting the relay out by the jacks in the enclosure. Get that to work first time, every time, when you put a logic signal into the driver transistor. Get the bugs fully out of one channel, then make the others. After that, you know that ANY logic that drives the relays will work. Then make the controller.

[tmichutka]

RG, you are the best!  I could never possibly thank you enough!  Per your recommendation I'm definitely going to use perfboard for this project.  Although, since I'm interested in learning I went to Fred Meyer (local grocery store) today and bought some Muriatic Acid and Hydrogen Peroxide because I'm thinking of using 1 channel as an integrated LPB clone connected to normalling jacks (in case I want to use that channel for a different effect).  I'm just going to use GGG's PCB design since I've already built that one succesfully once (although this'll be my first attempt at making a PCB).

Also, in case you or anyone else are interested in the Coto Shielded Reed Relays that I bought, here's the link http://tiny.cc/lUDoZ.  This seems like a really good deal to me, they're $5 each and they're 12VDC DPDT (2 form C) Electromagnetically Shielded Reed Relays with a coil supression diode built in (Coto 7142-12-1011).  Obviously I'm gaining nothing by posting this, I just wanted to share since it seems like such a great deal.  They seem like the perfect relays to build RG's Remote Switcher with.  I already picked up 10.

Well, I guess the next step is to order the components and try and get 1 channel to work reliably with logic.  Thanks again for the help RG!  I'll update this post in the future to let you know once I get it working and post some pics too.

[R.G.]

Just off hand, that sounds like the ideal relay, excepting only that it works from 12V instead of 9. Wow.

I've never actually used shielded reeds - the last time I looked they were over $12 each in the 2 form C versions.

Great find.

[tmichutka]

Okay, dangit, I thought that would be my last question, but I have a couple more... I hope you don't mind.

First of all, I got the Reed Relays and tested them out but I'm confused about one thing.  Without hooking them up and throwing the poles I'm not sure what's a pole and what's a throw.  Right now 3 & 6 are connected, and 5 & 8 are connected.  Below is a daigram from the datasheet with all the info I do have (ie Pin 1 = 12V+, Pin 2 = GND, Pin E = Shield).



Second question I may have answered for myself already, but if I'm going to use SPDT switches for the remote control (they're actually cheaper than momentary stomp switches now) with a Resistor and NPN Transistor going to the Relay, would the easiest method for LED indication be to put a separate Resistor and LED in parallel with the Pole of the switch?  DPDT switches are pretty cheap too so I could also get one and just hook up ground for the LED.  Would that work better?