lylehaze

OK, sorry you're feeling sick.. I'll dump a bunch of information here for you to look over later if you feel better. Just in case you didn't read it already, there's a pretty good detail on the difference between an encoder and a pot in the WIKI FAQ. I don't know how to link it, but the question header is: "What's the difference between potentiometers, sliders, motor faders and encoders?" The parts of the difference that matter to you are: Encoders connect to 2 DIN channels, pots and sliders connect to 1 AIN channel. Encoders are "relative", so you don't need to motorize them if the setting is changed. but pots are "absolute". so if they're in the wrong position, software will have to adjust things. That last part is important to understand.. If the fader is at zero, but the setting is really at 50%, what happens when you move the fader? Does it "JUMP" down to 1, or does it wait for the fader to reach 50% before anything changes? MIOS has some choices to help with this, but each choice is a different compromise. Unless that are motorfaders. then MIOS will just move the slider to whatever position it needs it moved to. Finally, getting back to the faders you have chosen: They offer AOUT only if you buy them with that option. This is the ONLY option that MIOS is built to handle already.. But I looked further, and there is one little problem. These faders run on 3.3 Volts, and the AOUT will swing from 0 to 3.3 volts as you slide the thing around. MIOS normally expects 0 to 5 volts, so you'll need to "scale" those inputs in software. (could also be done in hardware, 1 op-amp per slider) The SPI option is there too, in all models, It is "absolute", telling you the exact position of the slider with 8 bits of accuracy (256 steps). It is not "relative" like rotary encoders are, so even if you adapt SPI into encoder signals, you'll still have a problem with absolute/relative conversion. I don't see any way around that. The UART option is not documented in the datasheet, but I'll guess it's as absolute as the SPI is. So on to the bottom line: You can make this thing into an analog slider (pot) by ordering the sliders with the analog option, then adding a op-amp and a few resistors to scale the voltage to 0-5 volts. These signals will work great with an AIN module, and MIOS will treat it like any other pot/slider. You may be able to skip the op-amps and scale it in software, I can't say for sure. (I know how to trick the PIC into it also, but I'm getting off-track) You can challenge yourself to make an SPI to quadrature adapters with small PIC chips, and convert the voltage from 3.3 to 5 in the same step. But it won't be quite right, because MIOS assumes that all encoders are incremental, and these sliders will always be absolute. You can challenge yourself to make UART to quadrature adapters with small PIC chips, and convert the voltage from 3.3 to 5 in the same step. But it won't be quite right, because MIOS assumes that all encoders are incremental, and these sliders will always be absolute. That's all I can offer you. It' not what you wanted, but it's all out there. I really hope that others will correct anything I have said wrong. Feel Better, LyleHaze

I didn't read all six pages of clockbox forum posts, but it looks like the current schematic uses a DOUTX1 for the eight LEDs. If you use a DOUTx4 instead, you'll have three extra groups of eight outputs.. Use each one for one 7 segment display, and you should have control of the display from your core. Of course you'll set the resistors on those DOUT groups to whatever your display segments need. From the code side, there is a 3 digit BPM display on the LCD. So you can either find the code that puts those together, OR you can reverse-engineer the three digits from the display routine like this: The LCD is programmed with ASCII characters. the codes for Zero through Nine are actually 48 to 57 (decimal). SO if you grab the ASCII codes, then subtract 48, you'll have numbers zero to nine. Finally, you'll need to "draw" how you want each number to look on the 7 seg display, and write a small subroutine to convert 0-9 into those shapes. Since 7 segments fits into a byte, it's pretty simple. Finally, every time the BPM is updated, you'll use the above "stuff" to get the correct codes to send to the last three DOUT registers. Yeah,I know. That's easy for ME to say. Really it's not too bad. A good chance to spruce up your programming chops. Have Fun, LyleHaze

Hey Y'all. (That means "everyone" in Southern-Speak) Here's a link to the fader he wants to use. It is NOT a simple fader.. http://www.infiniumtechnologies.com/downloads/20090617_2/download It appears to have Analog, SPI, and UART options. MIOS / MB64E Expert opinions requested.

Ikea, Sorry you're having a rough time with this. I hope my comments are helping and not making it more difficult. I'm really _not_ trying to give you a hard time. I will offer one more suggestion: Because you are asking about a very specific part, it would really help you to get solid answers if you post the link to the datasheet of the part you want to use. Until I read that sheet I was sure that I was right and you were wrong. Once I read it I gained a whole different perspective. :-) In any case, good luck sorting this, and I'll help if I can. Maybe I'll be helpful and post that datasheet for you.. if I can figure out how. Us hardware types are pretty dangerous with URL's, you know. Good Luck, LyleHaze

I was wrong.. You taught me something today. Those really are linear encoders! Bad news, the "regular" encoder inputs to the MBox expect what is called "Quadrature encoder" signals. That's just two square waves that are 90 degrees out of phase. The device you showed me offers a choice of: UART, SPI, or Analog. You can teach a PIC to read any of those, but none match the "regular" encoder inputs we are ready for now. UART is a serial port.. SPI is "Serial Peripheral Interface", kinda like I2C but different. there MAY be a SPI driver in MIOS..?? Analog would attach to an AIN module (oh, now were back THERE again.) The other choice, (not so bad really) would be to add a small extra PIC just to translate UART or SPI into quadrature signals for MIOS. Sounds crazy, but it might be the easiest way to open the door for these cool faders. As always, I might be wrong with all of this. I've been wrong once today already. :-) Maybe a MIOS GURU can tell us if MIOS does SPI. Still, very cool faders! Good Luck, LyleHaze

Looks sweet. a really nice project. There is some confusion about encoders, pots and sliders.. An encoder is a knob that can turn endlessly without reaching stops. An encoder connects to a DIN board and uses two inputs. Normally all the encoders are grouped together. A pot is a knob that has stops. It is a resistor with a sliding tap point. These are measured in Ohms, or more likely Kilo-Ohms (thousands of ohms, or just K). Each pot requires one analog input(AIN). A slider is a pot that goes in a straight line, not a knob turning. It also requires an AIN input Pots and sliders can be "scaled" to be linear (even distribution) or Log (tapered to match our hearing). So if you're using pots or sliders you might want to watch that. Usually volume controls are LOG and other controls are linear. When you speak of I get confused. encoders and 10K linear faders are different creatures.. you may be able to change the software to replace one with another, but you cannot connect the 10K faders or pots to a DIN board. These will require some sort of analog input to work. Or maybe it's me that's confused. This is very possible. :-) Ouch.. It's too early in the morning to be thinking. LyleHaze

Looks interesting, _most_ electronic drum triggers are that I have seen are piezo. They actually generate a brief voltage spike when struck. They are force sensitive, but the voltage spike is very brief. These sensors you are looking at are resistive, and of course can detect how hard you are pressing continuously. If you're just experimenting, I recall seeing something very close to these that was homebrewed. Make the contact patterns in the same way, 2 layer or 1 layer interleaved, then use a bit of the black conductive foam that chips are often shipped in. As you compress the foam, the resistance across the traces should drop a bit. You could even "fake" the contact surfaces by using stripboard above and below the foam, with the contact traces facing the foam. Regarding feeding these to an analog input, there are "suggestions" in the MChip documents that voltages being fed into the ADC have as low an impedance as possible. The datasheet describes more details, followed by this in bold print:The maximum recommended impedance for analog sources is 2.5 k?. I don't know if the commercial device you are looking at is buffered, but the hack I described would need to be. Of course that would also allow you to "tune" with gain and offset to your liking. You MAY be able to see SOMETHING if you connect directly without any help, but the resolution and accuracy of the results will be less than ideal. Have Fun, LyleHaze

Trigger to MIDI conversion has different requirements than regular analog to digital conversion. You might be better served with a project tht is designed exactly for that. Here's a popular choice:http://www.edrum.info/ Have Fun

It is a nice, simple, clean design. I wonder if the side storage areas are the right width for adding rack rails? LyleHaze

SSP changes his nick to "Ikea". 8)

Oh, that is a horse of a different color. Binary values, each bit controls a different filter. More controls from less registers. You said: To keep this simple, I'll translate everything to decimal Bit 0 has a "value" of 1 Bit 1 has a value of 2 Bit 2 has a value of 4 Bit 3 has a value of 8 Add up the "ON" bits and you have the result to send to CC8 Google "Binary Numbers" for more on how this works. SO if I send a CC* with a value of Zero, all filters are OFF Value 1, filter for voice 1 ON, all other filters OFF Value 2, filter for voice 2 ON, all others OFF Value 3, Filters for 1 and 2 ON, all others OFF (1+2=3) Value 4, filter for voice 3 ON, all others OFF Value 5, Filters for voices 1 and 3 ON, all others OFF (1+4=5) Value 6, Filters for voices 2 and 3 ON, all other filters OFF (2+4=6) Value 7, all three voice filters ON, Filter for audio in OFF (1+2+4=7) the next eight are identical, except with the audio IN filter ON. It's not that confusing if you see the bits in binary. Let's count to 15.. 0=0000 1=0001 2=0010 3=0011 4=0100 5=0101 6=0110 7=0111 8=1000 9=1001 10=1010=A 11=1011=B 12=1100=C 13=1101=D 14=1110=E 15=1111=F So each bit controls 1 filter, the bit on the RIGHT is Bit 0, the bit on the LEFT is Bit 3. For each bit, 1=Filter ON, 0=Filter OFF It's like being in "The Matrix". Once you see numbers in binary, you don't need a translation. Have Fun, Neo

I'm guessing, as I've never built a Sid. But USUALLY, in the MIDI world, if you are controlling a toggle with a CC number, a value of 63 or lower will turn it off, and a value of 64 or more will turn it on. But most people just use zero (off) and 127 (on) If you're curious about why we use those numbers, try seeing those numbers in binary. It's a seven bit number, with a range of 0 to 127.. The MSB (Most Significant Bit) is OFF below 64, and ON at 64 and above, so they just use that MSB to control the feature. Have Fun, LyleHaze

One more for "interested also" I'm looking at building/programming both ends of the link: One end that reads the levels and outputs a MIDI data stream, and the receiving end that will turn that data stream into a bunch of flashing LEDs. For my own purpose, of course I want to integrate this with MBMixer. One app would be a real, flashy meter bridge. Another would be (maybe) a Java app and a PC based meter bridge a third possible app would be a Volume AGC circuit for television. (no LEDs needed) I don't yet know if the link will be standard MIDI or CAN. I see need for very high bandwidth and possibly long distances for this app, especially if mated with MBMixer. I just happen to have an 8X8 bi-color LED matrix on a pic, connected to this laptop right now. So development is already "in progress", but I won't predict how soon until have something to show for it. I'll probably just meter the outputs of MBMix for the proto work. Until I have nailed down an exact MIDI protocol, I'll gladly consider any "standards" out there that would do what I need. The less standards we have, the more valuable they are. LyleHaze

DJJ: Glad I could help. As usual, sometimes I give too many details, sometimes not enough. If I get feedback I'll make adjustments as needed. I try to describe WHY each test is being done. It is my hope to teach HOW to troubleshoot, not just how to get out of the problem of the day. TK: Lately I have been a little more active in the troubleshooting forums. If I step out of line, please feel free to tell me. I have no problem making adjustments if requested. Nice adjustment on the build guide. Sometimes it's the little things that make a difference. (That's what she said!) And they all lived happily ever after.. LyleHaze

It's beautiful man, nice work! But the picture from the inside, looks like a bad dream I had one night.. wires running everywhere!!! Nice Project. Lyle

Sorry, sometimes I get ahead of myself. Ill try again. From the beginning, With no power to the core, check the resistance of R6. It should be near 1.2K (1200 Ohms) With no power to the core, check the resistance of R5, it should be near 5.6K (5600 Ohms) Resistance tests are done with each meter lead on each end of the resistor, and the meter set to Ohms. These resistors are part of the PIC side of the 6n138 optocoupler, so I thought it would be a good idea to verify that they are correct. When I say "near", I mean within 5%. They do not have to be exactly right. If you're building from a kit, I'll assume that R6 is between pin 6 and +5 volts, and that R5 is between pin 7 and GND. If you hand-wired your core, please check those details too. Now on to voltage testing. You said that you only have about 4.45 volts at pin 26 of the PIC. To find the cause, we need to eliminate one possibility at a time. To begin, put everything back together, hook up the MIDI Cables and power it all up. Connect the ground lead of your meter to MIDIBox ground. (The tab of the 7805 is handy for this) Set your meter to DC volts, some scale above 5 volts (probably 20). Check the voltage at pin 32 of the PIC. It should be very close to 5 volts. If it's not, either your power supply is bad or your meter sucks. Fix the power supply or change the batteries in your meter. (don't laugh, it happens all the time) Now check the voltage at pin 26 of the PIC. You said before that it was 4.45 volts there. Is it still near that voltage? If so, proceed to the next step. If it's closer to the voltage you got from pin 32, your problem has magically disappeared, test your core. Now we have determined that the voltage at pin 26 is a little low. We need to determine the cause. because a meter reads an "average" of a fast moving digital signal, we need to make sure that what you're seeing is not really unexpected MIDI messages coming in. So please unplug the MIDI cable connected to the MIDI IN jack of the core. So now you have the optocoupler IN, but no cable connected to CORE MIDI IN. with the power to the core ON, check the voltage at pin 26 again. If it is very close to full voltage, then either you were getting MIDI messages, or you have a bad MIDI OUT from your PC. If the voltage at 26 is still a bit low, we will keep looking for the cause. power the core down, remove the optocoupler. power back up. check the voltage at pin 26 again. If it is back to near 5 volts, then we have determined that the optocoupler was pulling pin 26 down even with no signal attached. replace the optocoupler. If the voltage at pin 26 is still low, even with the opto out, your PIC has been damaged, order a new one. As always, I offer help for free, and it comes with a money back guarantee. Good Hunting, LyleHaze

That was rude.. Sorry. Just having fun. Pin 26 is where MIDI IN enters the PIC. That pin is an input, which means it's probably not the PICs fault (we'll get right back to that in a minute) It is normally pulled up to +5 volts by resistor R6. It is pulled DOWN by the optocoupler during an incoming MIDI signal. we'll check that in a moment. SO let's tear this down: verify resistor R6 as 1.2K from 6n138 pin 6 to +5 volts verify resistor R5 as 5.6K from 6n138 pin 7 to GND Unplug the MIDI cables, check the voltage again. Still low? Then it's not a bad MIDI OUT signal from your PC interface.(or unexpected MIDI traffic) Turn off all power, remove the (hopefully socketed) optocoupler. RE-apply power. Is the voltage at 26 still low? I'd be surprised. Most likely it's fine now, so buy a new optocoupler. If it's still low with the opto removed, pin 26 of your PIC has been damaged. replace the pic. (but I doubt it will be that) Let us know what you learn along the way Good Luck, LyleHaze

Dude, I thought you were taking a break.. ;-) Yes, the 4.45V on pin 26 is a clue. Any more questions?

Yes, the blacklist is still being updated, as we find more challenging tests. An excellent option for the DIY'er is the GM5, or the GM5X5X5. Both have been build by a bunch of people here, and bulk orders have been done (and may still be available..??) We even have the option of changing the names of the ports, if you add a small ROM chip. More info available if you search "GM5" on this site and the WIKI Have Fun, LyleHaze

OK, I'll throw a bit more in, maybe it will help you a bit: Making a "MIDI" device needs only a serial port, like most PIC chips have built-in. Making a "USB" device requires a LOT of low-level programming, that you might not want to get involved with. So if you want simple, make a MIDI device. Then buy a USB-MIDI converter, or build one from a kit like the GM5. Hardware to build: If you are allowed to build from kits, you need a core and a DIN for your project. Add a GM5 for the USB to MIDI interface. The projects already mentioned have more stuff, like encoders and sliders and stuff, but you can leave out any parts you don't want. If you're not allowed to use kits, you can build it from scratch. See the schematic drawings at ucapps.de for the PIC CORE and DIN modules. The parts you'll need are a PIC chip, crystal, optocoupler, shift register, some caps and resistors, DIN connectors, and a power supply with +5 volt regulator, It will take some time, but there is a LOT of testing and troubleshooting info along the way. *note: If you do not own a PIC programmer, you should buy your PIC chip with the bootloader already installed. There are two shops that offer those at a good price. Programming: You mentioned that you don't do much soldering. How about programming? At the easiest level you can modify the code for an existing project like the MB64E to make the buttons send whatever MIDI messages you want. If you want to be more involved, you COULD develop the entire application from the start, and not even use MIOS. That would be a lot harder, but not impossible at all. IF you try to do it all yourself, you'll also need a PIC programmer like the PicKit2, so you can flash your code into the PIC chip. If you work with MIOS, there is a bootloader already installed, and you can re-flash the chip by MIDI. Whatever you do, leave yourself lots of time to get it done. It always takes longer than you expected. I hope this answered a few questions.. LyleHaze

Hello WingMa, So far, the STM32 core is not ready for the public. It is still "Beta", so the PIC core is your best choice. I have never built with Motorfaders, so I cannot help with that. You will find that this is a GREAT site for people who want to learn more about how to make these projects. But it may not be so good if you are not interested in learning how. You need this by Monday? Stryds troubleshooting rule #1 may apply: Maybe a good plan would be to start with an existing project, like the MIDIBox LC. Once you have that working you can explore the software and learn about how it works a little better. This is going to take some time. Your project will not be completed in a week. But the satisfaction of making it yourself, and learning more about what goes into these things will be a great reward. Have Fun, LyleHaze

This is excellent news. You now have a tested path between your PC and your MIDIBox. Time for a big celebration!! re:sorting the display: There are hundreds of threads here from others who have had display issues. Search the forums, search the WIKI, you'll find every possible problem and each path to solving it. Some ideas include; wiring connections, setting 4 bit or 8 bit mode, wiring connections, It sounds like your power is connected properly. It might be a wiring problem. :-) One other suggestion, try loading an app that is already tested, perhaps midimon. That way you can separate hardware problems from code problems. You can always load different apps later once the hardware is working. Congrats on getting it running! now it gets better every step of the way. Have Fun, LyleHaze

TK is right, it sounds like you have a software "MIDI-THRU" that's screwing with you. Also, to clarify: The "loopback" test that I described does NOT include the MIDIBox, just a cable going from GM5 MIDI OUT to GM5 MIDI IN, I figure we get all that sorted before we put the MIDIBox into the picture. First, check MIDI-OX and remove the link from IN to OUT in software. Test by sending with NOTHING CONNECTED. what you send out should NOT come back. Then add a MIDI cable from GM5 Out to GM5 IN, send again, what you send out should COME RIGHT BACK, but on;y when the cable is connected. Repeat for each cable as a cable test. No doubt, you're making progress! Good Luck, LyleHaze

Which is which? I'm confused. Do you have a 110 volt transformer or a 100 volt? Your space echo power supply has a jumper inside (Transformer primary) that can be set for 110 or 117, so set it to whichever is closest to what you'll be feeding it. Have Fun, LyleHaze

This is back to you for an answer. You will have to change code no matter what. YOU said you wanted to use 7 segment displays. YOU said you need the days of the week. Can you display the letters you need on seven segment displays? Since I don't speak french, I don't know what letters you need. You can light any combination of the seven segments, those are the only options available based on the information YOU gave. There are 128 possible combinations of seven segments, including "all off". You can use any combinations you want, but if you can't display it in those seven segments, then you will have to change one of the requirements you have given. Here's a 2 digit 7 segment display (plus decimal points) Is this enough to display each day of the week? Also, you said you have 12 digits, but the display you are replacing seems to be using 16 digits, not including slashes and colons. You'll need more digits, or less information, or a display that changes or scrolls the information. Wow, I actually got an image to show! Have Fun, LyleHaze