Diode Switching

[Guest]

Hi guys, it's me again with yet another question... I hope you all can be as helpful as you were with the rest of my questions [:)]


I'm looking to install a Diode switch in my organ as an alternative to Syndyne and Electromechanical switching... I've seen someone post around before saying it was easy to figure out and make, albeit time consuming. I've talked to my local organ tech, and he has said pretty much the same thing. Now that I'm home for Christmas Break, I'm looking forward to hopefully getting a matrix made... But I was wondering if my research thus far is correct!





That's what I've come up with so far, is that the basic jist of a diode matrix? If it is, what resistor type should I use? I've built a few prototype models, and the only way I can keep the circuit from frying the resistor+diode leading to the stop control is to use some hefty wire-wound 10 watt resistors... Certainly that's not the only solution, is it?

[FrenchHorn8]

Re: Diode Switching

You'll definitely want to use a transistor or darlington to drive the valve magnets.  That way you can use a 1Kohm+ resistor.

[FrenchHorn8]

Re: Diode Switching

<span class="Apple-style-span" style="font-family: Verdana; font-size: 10px; white-space: pre;">
</span>

[davidecasteel]

Re: Diode Switching



I am far from an expert on this, but that circuit just does not seem right to me. I thought diode switching was based on back-biasing the switch diode to cut it off and I don't see that happening here. Also, I'm not clear on whether the Stop Switch is closed to enable the stop or open. When the Stop Switch is closed and a key is pressed, it enables a direct circuit from the +15v through the resistor back to ground, and depending on the resistor value that could be a lot of current. I don't think that is the way it should work.</P>


With the Stop Switch open, when a key is pressed it puts +15v through the diode and resistor through the magnet to ground, and the combined resistance of the resistor and magnet will cause a voltage drop of almost 15v (the diode resistance will be very small, since it is forward biased). To turn the diode switch OFF requires that it be reverse-biased and that means that a voltage equal to or greater than that at the anode must be impressed on the cathode, and that is not happening in the circuit depicted.</P>


I guess I'll have to let someone more educated than I figure it out.</P>


David</P>

[FrenchHorn8]

Re: Diode Switching

[quote user="davidecasteel"]

When the Stop Switch is closed and a key is pressed, it enables a direct circuit from the +15v through the resistor back to ground, and depending on the resistor value that could be a lot of current.  I don't think that is the way it should work.</p>

[/quote]</p>

That actually is the way it works.  Just a way of bypassing the magnet.  There should be a transistor so the resister value can be higher, therefore eliminating high current.</p>

[Guest]

Re: Diode Switching



That's awfully inefficient plus a transitor would be cheaper than a wire wound resistor. Tou can use an IC like the UDN2981 or ULN2803 depending on polarity, each package has 8 power transistors plus the reverse diodes. </P>


Al </P>

[davidecasteel]

Re: Diode Switching

[quote user="FrenchHorn8"][quote user="davidecasteel"]


When the Stop Switch is closed and a key is pressed, it enables a direct circuit from the +15v through the resistor back to ground, and depending on the resistor value that could be a lot of current. I don't think that is the way it should work.</P>


[/quote]</P>


That actually is the way it works. Just a way of bypassing the magnet. There should be a transistor so the resister value can be higher, therefore eliminating high current.</P>


[/quote]I'll take your word for it, but it doesn't fall into the category of being a "diode switch" as I understand it. It seems like a very ineffecient way to use DC power, too. What's wrong with just using the Stop Switch to provide the 15v to the bus that feeds the key switches? Then only the keys being played provide a drain on the DC power. </P>


I'm not criticizing--I already said I don't know anything about it--but I really don't understand this circuit.</P>


David</P>

[Guest]

Re: Diode Switching



An organ relay normally uses a multi pole relay activated (key relay) when you depress a key. There are 61 of those for each keyboard. Each pole goes through a 61 or 73 pole relay (stop selection or stack) and then to the pipe magnet. That's the way Wurlitzer did it in 1920. It still is used with variations today. I believe that was Hope-Jone's invention. You can eliminate the key relay by using diodes tied to the key contacts. A more modern way is to utilize technology: The keyboard keypresses are converted into a serial digital data stream (Midi) which then goes through a midi relay which processes the data to add stoptab control. This Midi data then goes to a pipe driver board which converts the serial Midi data into parallel signal to directly operate the pipe magnets. It's relatively simple, inexpensive (compared to mechanical relays) and quite reliable when properly designed. With this data many functions like transposing, coupling, etc are easily implemented plus the cabling is greatly simplified. 2 or 3 wires instead of hundreds.</P>


Al </P>

[Guest]

Re: Diode Switching

[quote user="FrenchHorn8"]You'll definitely want to use a transistor or darlington to drive the valve magnets.  That way you can use a 1Kohm+ resistor.[/quote]
Ah! The transistor was the missing piece. I've been thinking about this for a while, wondering which combination of resistors would allow enough juice to operate the magnets, and still provide a sufficient ground for the stop control...

[quote user="Al"]That's awfully inefficient plus a transitor would be cheaper than a wire wound resistor. Tou can use an IC like the UDN2981 or ULN2803 depending on polarity, each package has 8 power transistors plus the reverse diodes. [/quote]
Believe me- I'd use a transistor over a wire wound resistor any day... Now that I know about it. [:P] And thanks for pointing out those two IC's to me. Although a printed circuit is beyond what I'm interested in (imagine diode leads sticking through holes drilled in plywood), it's definitely nice to know alternatives are out there.

[davidecasteel]

Re: Diode Switching



There are a number of different prepunched blank circuit boards available for creating specialized circuits. You'd probably want one with just circular solder pads around the holes and no connections. These things can neaten up a build very nicely. If you are going to use some ICs, there are also boards available that have holes and connections designed for mounting ICs and wiring them into circuits. Do a Google search on "prototype circuit boards" and you'll find lots of possibilities. Archer makes small boards that are sold by many Radio Shack stores, also.</P>


David</P>


This is one example of what I mean: http://www.circuitspecialists.com/level.itml/icOid/5915</P>


And here's what Radio Shack sells: http://www.radioshack.com/family/ind...goryId=2032265</P>

[FrenchHorn8]

Re: Diode Switching



I just built mine on breadboard from mpja.com.  Same breadboard as Radio Shack, but 1/4 the price.</p>

 </p>

[Guest]

Re: Diode Switching

The only reason I've been considering a drilled plywood base for the switching is that I've seen a Zimmer organ in town that has it set up that way... It seems like it'd be more stable too, a smaller risk of a component being pulled out of the breadboard or something. Are the components pretty secure once they are stuck in there?

Also, FrenchHorn8, do you remember where you got those cable bundles from? I've been using 24 gauge 2 pr hookup wire so far- using the 25 pr or 50pr (whatever it is) cables looks a lot neater.

[Guest]

Re: Diode Switching



Using IC's a trick is to mount upside down on double sided foam tape and wire wrap to the leads. You can also get wire wrap sockets and bvoards. Very easy and reliable. The cables in the picture look like 25 and 50 pair phone cable. The phone cable connectors are also available plus the cable often comes with connectors attached. Look for any old commercial phone systen for this stuff. I used it 30+ years ago. Looking around you can often get it for free.</P>


Al</P>

[Guest]

Re: Diode Switching

Do you think an IC would definitely be better in the short term and long run than using standard diode+transistor...?


If I do go for the non-IC route, would these parts work for my usage:

For the resistor

For the transistor

Sadly, I don't know the ins and outs of electronics that well, so power ratings and whatnot go over my head.

[Guest]

Re: Diode Switching

Well, I took a leap of faith and assumed, based on some small math, that those would be fine. And they were. [:)]


Gentlemen, we have a prototype!




I can't thank you enough for all of your help, all of you. I'll be sure to post a picture or two of the final product.