MRE

I dont remember if MB uses the LCD in 4 bit or 8 bit mode. Either way, in order to configure and start an LCD, the micro sends control charactors down only 4 bits. So, if the LCD displays anything at all, then it was properly set up. However, if its displaying strange charactors, then it is either A: rapidly resetting (usually caused by a CPU brownout [low core voltage]) Or B: the other half of the bits are mis wired somehow. Example: Micro connected to LCD in 8 bit mode (Plus 2 control bits). Micro will only use 4 bits at first to configure the LCD. This either is correct, or not. If there is a problem initializing the LCD, the screen will likey not show anything, and certainly not data that changes due to input from the user. If, however, the other 4 bits are misswired, then the LCD will get 'apparently' correct data. I.E. it gets a charactor code that corresponds to SOMETHING.. just not what it actually SHOULD be. The other possability is that during LCD initialization, it is set to one data rate, and the core actually attempts to talk at another data rate? Maby... I dunno about this one. So, double check the high bits on the connection. Check for brownouts in the core. Check that nothing is shorting the connector to the LCD. I doubt MIOS is configuring one data rate and talking at another. Try disconnecting all IO boards from the core, and power up just the core and LCD. Or try powering the LCD from another source if its VFD or other high current demand LCD.

I partially agree with stryder in that there is very little money saved in not building extra cores. They could be built on protoboard very cheaply, and a core kit isnt that much more. Building a backplane system and compact flash would be a lot more expensive than simply buying a few core kits. That being said: http://www.walrus.com/~raphael/pdf/compactflashOnPicArticle.pdf Right away you see that you might need some logic to get the flash bus to play nice with the rest of the Port C/D and E duties. Good luck with THAT!! Perhaps a different suggestion? : Why not write some special code for the bankstick? Granted, a MB64 and a SID may not interchange, but maby you can manage to switch MB64 and seq? You would have to figure a way to load a full config file into the PIC ram, but it may be easier than via CF. The technology is 'built in' and the code is already half baked.

and Stryder says "booo-yaaah to yer grandma!!" Exactly what he said. It helps that less and less often you see 'strange' driver ICs in the surplus market, meaning that most of the time whatever LDC module you get will work. Also, LCD modules often conform to one of two interconnect standards (pins in a row, or pins in two rows). Wacky pin configuratons still exist, so always double check the datasheet before making your cables.

Another good choice. Parts cost likely between that of a voltage divider and an opto. Dont forget .7 volt drop on the Diode, so thats actually 14.3 Volts in the equation. (wont make a lot of difference anyway. Note that TYPICAL does not mean MAX I ;) I dont know anything about those old organs. Is the key voltage 15 volts when the key is up, and zero when the key is pressed? or the other way around? (doesnt matter to the circuit, only curious)

Screaming Rabbit is right on the nose there. The Midimon is as simple as it gets. However, its a monitor or test/debug tool. Its not an 'instrument' on the midibus. You cant make sounds with it. Fortunately, Midibox is incredibly scale-able. The smallest Midibox instrument a person could build would simply consist of one knob, or one switch, or one encoder. (plus a few buttons for menu control) It is a core, an LCD screen, and a few buttons plus a _________ (your choice there). You would then need to choose the appropriate i/o module .. example: a knob would need the Ain board. Now, there IS the choice of modifying the code in order to directly wire the knob to the core, and thus avoiding the Ain circuit. For newbies sake though, buy the boad, or at least build part of the circuit. Now, finding one knob (or fader, or button) usefull.. is up to the user. The good news is that once ONE knob works, its a simple matter to add many more. It really IS just soldering some wires to the pot, hooking them to the Ain board and calling it a job well done.

I am seeing from your post that you perhaps think there is an LCD module (circuit) that would connect between the LCD screen you purchased and the Core module. I think some language being used must have confused you. Many people use the term 'LCD module' to refere to the LCD screen. The LCD you purchased should connect directly to the Core (assuming you create the proper wire connections). There is not a circuit board that is to be connected BETWEEN them. You SHOULD double check that your LCD screen is compatable with the Core (most are). MIDI soundcard: It may be best (easiest) to simply purchase a USB - MIDI unit. There are several with Midi In, out and thru. They go for prices well under $100 for a simple (no thrills) device.

The other option would be to use optoisolators. There are several that can handle higher input voltages and always give a clean 5 volt signal out. (cant remember p/n off hand) A lot less worry, but a bit more expensive than a voltage divider. So long as you are confident that your key voltage is stable and doesnt swing over, the voltage divider should be fine. If you question the stability of the PWS though, I would go for optos. Designer choice ;)

I know communications engineers (especially those from the stone ages) who are less concise with their terminology than you Stryd, but yes you are correct. Hats off. I generally try to remember to use RS232 or at least COM to refer to the com port of a pc (and thus, RS232, unless its a 'faked' port ex: via USB). Sometimes I still toss 'serial' around when in fact I ment (read: should have said) RS232. I almost always differentiate 422 or 485 however. Bad habits die hard.

In the interest of harmony (and a clean message board) I recant everything I said (courtesy of the Remove button) and agree totally with AudioCommander! (which, I did anyway) :P Yes, MIDIbox can control all sorts of fantastical things. It can be triggered by, and trigger many other magical wonders. However, if you want to build a "ROBOT".. get a Lego RCX. I got longwinded and off topic. I am still waiting for my Ones and Zeros Applejax. :D

If you are seriously concerned about the coding thing (sometimes even cut/paste can be a pain anyway.. not to mention there are some steps involved in just getting it all loaded in right), then I would sugest that you stick with the 'stock designs' as much as possible. That way, you can load the default configuration files and be done with it. The less mods you make, the less someone will have to help out configuring it just right for your set-up. You can always dual core if you need more Ins and Outs. Anything not being used can be tied low, and reserved for expansion later.

For hand made work, I use Vectorboard or similar pretty much exclusively. I would sugest that you try part 500/15 since it most closely represents a typical breadboard. So, you could create a test sample of the circuit on breadboard, play around with the smartest way to wire it up, then simply transfer it to the Vectorboard. These boards are GREAT for circuits with a lot of IC chips. If you choose another board, just think carefully about how you will build your circuit. Example being DIN and DOUT modules might be better served by a different Vector board. When making circuits with a lot of descrete parts (not IC chips) or that need a particular layout (such as control panels), I do the tedious point to point wiring on something more like the 500/19. Be carefull with heat on those individual pads. After a solder joint or two, they tend to come off the board. Regarding your Rectifier question: Are you refering to a full bridge rectifier? If so then, If you cant find it, then 4 diodes of equivilent current ratings will do. Look up a rectifier circuit diagram to arrange the diodes correctly. I dont know the ratings on the parts list off the top of my head, but the diode you stated sounds more than enough.

ok! I tried the example search Orwell "Animal Farm" -movie and got nuthin! WTF mate?! hehe..

ya know, I looked over the input schematic for the eDrum again, and its actually a pretty impressive circuit. It presents a 0 to 5 volt swing to the analog inputs on the PIC. It clamps the signal to not exceed 5 volts, thus making it 'friendly' to digital logic. So, you could use this directly with any digital input. There will be a switching range (governed by the input chip's 0 to 1 threshold) that you need to pay a bit of attention too.. so.. in otherwords, a small trigger on a pressure sensor might not trigger the digital input (say the signal is less than 2 volts). But take a look at P1, P2, etc.. this is essentially a sensativity adjustment. I would guess that in MOST cases, you need mearly build this circuit for each input, and tune Px so that the output swings over 2.5 volts when you trigger it with the minumum expected trigger. This will be rather LOW sensitivity compared to how you would set it on the eDrum. Thus, any trigger over your minimum setting will swing the opamp high, and output up to 5 volts. Even if you whack the trigger really hard, it will clip off at 5 volts, and not damage your input. For some types of input, you might need to experiment with the gain settings (R1, R2, and R3 on the first input). Get the data sheet on the LM324 before you go poking around.

XP on the mac is dodgey at best.. I did an install for a friend of mine, and of all things, it couldnt find the sound device (nor the batter monitor hardware for that matter). All I can say is.. do a diliberate forced install of device drivers (audio, midi, and whatever driver you are using for midi-box). Maby it will show up?

You should seriously check into how op-amps work. Particularly as used in threshold switching applications. In addition, I would highly recommend you look into general data aquisition sensors. Check over the E-drum schematics carefully. I know you wanted to bypass that whole idea, but you jumped away from it without even knowing what the designer was doing with it. Pay particular attention to the pad input circuits. You could just as easilly use the same input circuit for ANY type of sensor (provided you tune the resistors to match). Velocity data is primarily software based. E-drum would work on exactly the same electronics, with or without velocity data. Some sensor ideas: Piezo microphones - attached to opamps work pretty well for any input that creates sound (this is essentially the same thing as a piezo trigger like E-drum. The only difference is that the mic has more sensitivity to MANY forms of input (impact, audible, pressure, vibration), rather than just impact). Imagine that someone doesnt mearly have to sit on the couch, but need to make the couch squeek. Ummm.. how about a midi sex bed! heh... only makes music when you are.. umm.. 'making music' ;) I call it art! Strain Gauges - these are small pieces of thin film on which a set of resistive traces are placed. As the film flexes, the resistance changes. If you put this in as part of a Whyne bridge, it is incredibly sensitive. (Ive used them to measure microflex on 3 story tall mining trucks. The twisting is imperceptable to the human eye) So, this would be perfect for items that need to flex over large surfaces. For instance, you can put a postage stamp sized (or smaller) strain gauge on the CORNER of the stair, by the wall. As a person steps on the stair, that corner is going to flex ever so slightly. You will get a good solid trigger if the output from the whyne bridge is fed into an opamp. for more fun, a two axis strain gauge will be able to 'locate' where on the step the person put their foot, as the corner of the stair will flex both lengthwise and depth wise in relation to the foot placement! In the case of a couch or chair, mount it to a part that is relatively stiff i.e. part of the frame, but close to where the cushion attaches. You will get a definate response, without over flexing the gauge and breaking it. Good ole light sensor - Get the resistive type. It is a lil glass bubble with a piece of plastic inside. On the plastic is a 'race track' of wire that is responsive resistively to variations of light. Again, whyne bridge to op amp, and you can tune exactly what level change you want it to trigger on. A wave of the hand.. or a shadow cast on the wall (and thus the sensor) and you have your trigger. Even thermal sensors could be used as triggers.. perfect for that hightech fart gag. The person sits down, and warms the seat. At a certain threshold, the trigger fires.. when they get up in startled confusion, the sensor cools off and you can play that diarhea squirts sample you always wanted to use. Point is.. if you take the time to learn a lot more about data aquisition technology, and op-amps in particular, you can make just about anything an input. Trust me: look at the edrum again. It is only limitied in inputs because of the constraints of the pic. The input circuit itself will be VERY usefull to you.

Looking at the DOUT board, on the left of each 'bank' of resistors, is the ground point for that group of LEDs.. Technically, each of these 4 ground points is the same. So, if you were tie all of your LED grounds back to any ONE of these ground points, it would work fine. So, I would suggest a variable resistor in which one leg of the resistor is tied to the ground, and then all of the LEDs are tied to the wiper. That way, you have full control over the brightness of all the LEDs at once (connected to that board). Just make sure its a decent sized VR, so that it will handle the current. To make it easy (ie, you already have all your grounds on the DOUT board and dont want to muck it up again).. cut the trace on all 4 ground points on the board, then wire a series of solid core wire jumpers on the connector side of the cut to each of the grounds. Finally, insert a VR between the cut points of ONE of your cut traces. DO NOT cut the part of the trace that grounds each chip! (or the extension connector for that matter!) In fact, there is a bit of room on the far left of the board. Perhaps the next revision of the DOUT pcb could have these traces cut, with a set of holes, and a pad set for the pot on the left side of the board. That way, you can install a pot if you want, or simply install 4 jumpers if you dont?

Absolutely get the spec sheet on the pins. They 'generally' follow a standard, but some glass makers have their own ideas. Also, if at all possible, get a connector for the LCD.

I remember the data cables being very similar to midi cables.. and it looks like you are using those same holes.. nifty.. now if only you could do something with the platter switch :)

heh.. havn't tried either one. It was the hot air gun idea that got me thinking about the torch.. which we all know what happens with that one! The hot air gun method was demonstrated on Hackaday.com a while back though. Would think the toaster method would be a bad idea as it somewhat evenly heats the entire board, and chips. You want to heat the pins yet keep the chip itself as cool as possible. To get the all the pins hot enough to melt the solder enough to pull it up via a toaster, would mean the whole chip (delicate silicone transistors included) up to the same temp. sounds like a bad idea to me..

Well... yeah.. if you want to get rid of a bad part just snip it off! I do that all the time to repair boards.. but the topic was recovering parts from one device to use in another. When you dont care about the part, its easy ;) As to the air trick... I havnt used a can of air yet.. I have only done this method with a brisk puff from my own lungs. It sort of my last ditch effort, when the wick and the pump didn't do the job. Or putting a fat glob on the pin and then sucking it all back out didnt let it loose. It does work quite well. The repeated heating from the iron is far more dangerous to the part than a brisk blow of air, even if it were cold air.

Yeah thats a nice one. Like that you are using the original drive LEDs too.

yeah for fun I tried torching a board after this post... what a waste! hehe another idea: you need quick hands to do it, but I have had success with melting the solder one pin at a time and then while keeping the tip on the pin, hitting it with a quick strong burst of air. I usually blow on the hole from the component side of the board with the iron tip on the solder side. You could also attempt to blow on it with a quick shot of compressed air. Obviously, you need to pay attention to the blow out area.. IE where is that bubble of hot solder going to fly out too?

TK has put forth a substantial amount of effort in getting MIDBOX off the ground. He did it for his own personal use, and was willing to document it for everyone to share. He could have made money for the project, but has chosen not to. I would suggest that if you ask for money for work, that you do it in the following way: 1 - first and formest, you can only charge for your LABOR and MATERIALS. IE you can certainly profit off the time you spend building a box for someone, but you cannot profit off the time that TK spent designing it in the first place. 2 - To be fair, you should buy kits from the suppliers listed rather than build your own. Why? you will continue to benifit the community. In this way, you pay for your right to use the MIDIBox platform for your own needs (IE the needs of the person you are building it for). The suppliers have incentive to keep them updated. Also, money trickles in to TK's wallet. Its only fair after all! 3 - you detail out the bill to the person. IE The midibox hardware kit costs are clearly stated, and the website you bought the kit from is clearly stated. You should not 'mark up' the kit costs. Other costs are included (material costs for the box, faders and pots, lcds etc), and finally your cost for your own time and labor invested. If you do your best to be fair about it, you will realize that the production cost plus a small profit for your time equals more than an off the shelf unit. The only way to be fair about this is to point out that it is a custom unit fit for to the specific needs and desires of the person. The flip side is that YOU are responsable for supporting it through the life of the unit.

multipin components are generally hard to take off. I usually work slowly back and forth with the pump or solder wick. While heating one pin, push firmly with your thumb with the part. The idea is to 'wiggle' the part of, back and forth, by melting alternate ends of the part. The other tip would be to remove as much solder as you can, then hit it with a hot air gun or torch. Obviously with the torch, you burn the board.

Assume you will also need to cut 418 holes! ;)