Synapsys

Hi, It occurred to me while reading this thread (and many others here) that many MIDIBox builders could benefit from having a logic probe to perform troubleshooting on their projects. Most builders here cannot afford or justify the purchase of an oscilloscope. A logic probe is an inexpensive yet powerful tool for troubleshooting digital circuitry. Here is a link to one example of such a probe. http://kitsrus.com/projects/k24.pdf A search on the Internet for “DIY Logic Probe†reveals many other such cheap to build logic probes Although I am fortunate enough to own an oscilloscope I use my logic probe regularly when troubleshooting digital circuitry. Hope this info is useful. :)

The data sheet for the 18F452 recommends that the source impedance for an analog input should not exceed 2.5K. The higher the source impedance the longer it takes for the ADC to acquire a change in the input voltage. In addition, as the input impedance increases so does the offset voltage at the input pin. This offset voltage is actually an error voltage. In this application I do not believe this error voltage is very significant. I assume that TK optimized the code to allow enough time to acquire a change in input voltage based on the 10K pot. That being said I would think the 2K pots would be a better choice since the acquisition time would be shorter than with a 10K pot. The 2K pot will as mentioned draw more current (2.5ma per pot) and pay_c is right about making sure the power supply can handle this load in addition to the load of the remaining circuitry. :)

Hi Thorsten, I am a bit confused by your comment about the transmission speed. When you said "(thats also the case if you would use the internal peripheral)" were you referring to the MSSP? As I read the data sheet for the 18F452 it appears to indicate that using SPI mode and a fosc/4 SCLK rate the transmission speed is 10mHz with a 40 mHz oscillator. That means one byte would be transmitted in 1.25usec. And since the SPI uses an interrupt to indicate transmit / receive complete the processor is free to do other things during this time. I don't see how this is any different (other than a significantly higher speed) from using the uart for inter-processor communications. What am I missing here? :-[

Hi Dave, Sorry I confused you. I guess I left out a very important point in my reply. :-[ When I referred to the core module schematic's use of RB5 I meant to say that in order to program the PIC for use on the core module you must set the configuration so that the RB5 pin is an I/O pin. This disables the LVP capability of the chip. After this is done you cannot use LVP on the chip again only HVP. The only alternative is to not use RB5 as an I/O pin. This would require changing how the LCD on the core module works and thus the MIOS LCD driver as well. By the way, does anyone know why MIOS does not support 4-bit data interface to the LCD? Or maybe it does and I just have not seen any documentation on it. This would free up 4 more I/O pins on the core.

Hi Dave, Sorry I confused you. I guess I left out a very important point in my reply. :-[ When I referred to the core module schematic's use of RB5 I meant to say that in order to program the PIC for use on the core module you must set the configuration so that the RB5 pin is an I/O pin. This disables the LVP capability of the chip. After this is done you cannot use LVP on the chip again only HVP. The only alternative is to not use RB5 as an I/O pin. This would require changing how the LCD on the core module works and thus the MIOS LCD driver as well. By the way, does anyone know why MIOS does not support 4-bit data interface to the LCD? Or maybe it does and I just have not seen any documentation on it. This would free up 4 more I/O pins on the core.

Cool on the solar lamp. It must have light in the UV band. As long as the traces are intact and not shorting to other traces there should be no problem.

Cool on the solar lamp. It must have light in the UV band. As long as the traces are intact and not shorting to other traces there should be no problem.

Hi SounDuke, The white paper on the board is there to protect the photoresist during shipping. It MUST be removed before exposing the board. The light used to expose the board must be an ultra-violet fluorescent type. An ordinary incandescent lamp will not work. Take a look a this web site for detailed description of the process. http://www.mgchemicals.com Happy etching...

Hi SounDuke, The white paper on the board is there to protect the photoresist during shipping. It MUST be removed before exposing the board. The light used to expose the board must be an ultra-violet fluorescent type. An ordinary incandescent lamp will not work. Take a look a this web site for detailed description of the process. http://www.mgchemicals.com Happy etching...

Hi Dave, I do not have any advice on the JDM as I have not built one. But before you build a low voltage programmer consider that the RB5 pin on the 18F452 cannot be used as a I/O pin while low voltage programming mode is enabled. The core module connects RB5 to a data pin on the LCD so loss of this pin would mean a programming change of some type (either a configuration or MIOS change I am not sure which). I looked at the site you posted and there is also a high voltage programmer schematic there. Maybe that will work for you. :) Regards, Synapsys

Hi Dave, I do not have any advice on the JDM as I have not built one. But before you build a low voltage programmer consider that the RB5 pin on the 18F452 cannot be used as a I/O pin while low voltage programming mode is enabled. The core module connects RB5 to a data pin on the LCD so loss of this pin would mean a programming change of some type (either a configuration or MIOS change I am not sure which). I looked at the site you posted and there is also a high voltage programmer schematic there. Maybe that will work for you. :) Regards, Synapsys

Hi pilo, Yes the circuit in the schematic works very well for guitar. I use it for electric as well as acoustic with bridge based pickups. The OPA2604 is the key to this circuits success. It is very low noise and has very high input impedance (necessary particularly for acoustic pickups). My fx processor project I mentioned is completely analog. It contains a tube preamp (can be switched in/out of signal path), some filters for cabinet emulation, a four band EQ and a compressor/limiter. I used the (relatively) low voltage tube circuit that can be found at www.PAIA.com. It works pretty good but I think if I were to build another I would want to drive the tube with a higher plate supply voltage (not recommended for most hobbyists since the higher voltage is dangerous … even deadly :o). The RC4580 is a good replacement for the NE5532 but watch the supply voltage. The RC4580 is max'd at +-15 volts whereas the NE5532 is max'd at +-22 volts. The TL072 chip is good for line level signals. I used the TL074 for the EQ circuit. It is also good for peak detectors but I would avoid using this chip as the output driver. My experience is that it does not perform well driving long cables. Psytron, There is no component at the point you indicated. The input jack is a shorting type (thus the up arrow). The ground plane indicates that the signal path to the capacitor on the PCB is surrounded by the ground plane to reduce noise. I use PCB mount jacks to reduce cabling that certainly all MidiBox fans know is a real pain. :'(

The 22BP caps are bi-polar electrolytics (DigiKey # P1177-ND made by Panasonic)

I contacted the US rep for Song Huei and found out the MVA02-1101 series motor faders cost 46.50 USD each FOB Taiwan. The catch is the minumum quantity is 1000 units!! :(

Hi LO, Here is a schematic for a preamp that works well with a guitar. It can be built for far less than $65 (under $15 USD for parts). I used this circuit as the front end of a guitar effects processor I built several years ago. The circuit requires both positive and negative supply voltages. I built a power supply for the processor but I do not recommend that due to the danger of working with mains. For a DI box I believe two 9-volt batteries as showm on the schematic would work okay. Or you could use two 9-15vdc wall warts if batteries wear out too fast. Hope this is useful... :) http://synapsysllc.com/guitarpreamp.htm

You are right about the availability of bulk resist that can be applied to plain copper clad boards. I have never tried out this process myself but here is a link to someone who has. http://w1.859.telia.com/~u85920178/data/pcb2.htm

psytron is right about the toner. I also have an HP LaserJet and I bought an extra toner cartridge that I use just for PCB work. It has lasted over two years and I do a lot of PCB work (mostly analog). Use of an Inkjet to produce the transparencies is not advised. Inkjets do not produce truly opaque images, which means the UV light will penetrate the image resulting in poor quality traces. Speaking of quality, all laser printers I have seen have a setting that controls the "darkness" of the printing. The default for this setting is usually medium. It is best to change this setting to maximum darkness, print the transparency and set it back to medium. About the 'more chemicals', the only additional chemical is the developer and it is not at all noxious. It has no odor. Care is required to keep the developer off things like clothes since it will tend to bleach them. About etching; I use sodium persulfate instead of ferric chloride for all my etching. It also has no odor or noxious fumes like ferric chloride does and it is clear so you can see the etching progress without removing the board from the tank. It is also important to handle an unexposed board in the 'dark'. With the MG Chemical boards I use a small (40 watt) incandescent light while working with the unexposed board and have had no problems. Another tip is to print the PCB layout as a mirror image and then turn it over so the layout is correct. Doing this places the toner side of the image directly against the board. If you don't do this the thickness of the transparency allows the UV light to sneak under the image and the results are poorer. One final thought; use the highest density laser printer you can afford. Mine is 1200 dpi and the results are great. I have used 600 dpi in the past with good results. I would assume that if you had the transparencies done at a print shop using 4800 dpi the results would be even better. happy etching... :)

I use the "Laser" transparency sheets in my laser printer all the time without heat problems. I have tried the "iron on" technique a few times with fair results on simple boards. However, I invested a few bucks in equipment and now used photographic techniques and get excellant repeatable results. The method is quite simple. You print a positive image of the PCB layout on the transparency and place it on a pre-sensitised blank PCB and hold it tight to the board using a piece of glass. Then the board is exposed with an ultra-violet fluorescent light for 5 minutes. Once exposed the board is placed in the developer and in seconds the exposed portion of the resist dissolves away leaving the unexposed portion on the board. The board can then be etched in the normal fashion. I use MG Chemicals products found here http://www.mgchemicals.com Using this technique I can easily produce board with 15 mil (0.38 mm) traces on a 25 mil (0.64 mm) grid and can run traces between IC pads spaced 100 mil (2.54 mm) apart. Although this method is a bit more expensive than others it reliably produces extremely high quality boards. Hope this info is useful.... :)