JDM Programmer not detecting after Nigel Tutorial 1.2??

[Sceadwian]

If you still want to go through the trouble of it, I supposed you could buffer the incoming RS232 from a real +12/-12 supply, I'm guessing it's a failure of modern machines to provide proper voltage levels and the needed current draw? What would you need for that? A +12/-12 supply, and a transistor as an emitter follower?

 

[Nigel Goodwin]

If you still want to go through the trouble of it, I supposed you could buffer the incoming RS232 from a real +12/-12 supply, I'm guessing it's a failure of modern machines to provide proper voltage levels and the needed current draw? What would you need for that? A +12/-12 supply, and a transistor as an emitter follower?

Easier to build a proper programmer, the JDM doesn't even use the serial port, it just uses some of the handshake lines from the serial port and bit-bangs it, like parallel port programmers do. The main real difference is that the JDM attempts to supply the 13V Vpp voltage from the handshake lines, which is where the problems start.

I've always considered it a very clever piece of design, but that's just it - it was designed to show how clever the author could be, and not to produce a reliable programmer.

 

[Tightwad]

Am I seeing it right that the PicKit2 is just a module that plugs into a board that holds the chip? So I would need to purchase the PicKit2, and a board, or build a board?

Can the Junebug do 16F628A? 16F648A? 16F88?

 

[blueroomelectronics]

Yes to both. You generally need 3 pins to program a PIC in circuit.

 

[Nigel Goodwin]

Am I seeing it right that the PicKit2 is just a module that plugs into a board that holds the chip? So I would need to purchase the PicKit2, and a board, or build a board?]

It's a ICSP, it programmes the PIC in circuit - you just need to design the target board suitably. Or, you can make a target board with nothing but a socket - plug the PIC in the socket, program it, then take it out and place in the target board.

Can the Junebug do 16F628A? 16F648A? 16F88?

Yes, it does a huge range of devices, and new ones get added all the time.

 

[Tightwad]

Ahhh, I never really understood the ICSP part. I like the idea of either In Circuit, or a Zif based small board, or both.

Where can i look for details about how to incorporate...pins needed, isolation, etc?

 

[blueroomelectronics]

Microchips site has an ICD2 connection guide. If you're new to it just use a PIC with more I/O than you need so you can spare PGC, PGD & MCLR.

 

[Tightwad]

Thanks for the help. I am so new to it that I don't know why I need pretty much anything...but it has been fun to learn so far. My goals for this project are not that big...take an input and override a function on my Motorcycle for a set amount of time for instance. Mostly I wanted to learn.

I think the 16F648A, 16F628A and 16F88 will be all the chip I need, as my needs are fairly simplistic. Of course once i finish I will want to expand upon it, which is why I am using the PIC instead of doing it via all hardware and timers.

If i want to use the Junebug to program a chip, i will still need a jumper board, right? I will look at Microchips site to learn more about the ICD2 connections

 

[blueroomelectronics]

Yes the Junebug has a 2x5 header for a ribbon cable jumper to a target PCB. All my kits use this same header so if you look at the manuals you'll see how they connect to target PICs. Also the Inchworm Quick project on my download page shows how an ICD connects.

 

[Sceadwian]

Right Nigel, as I said, if you supply the target board with power directly.. Don't the the problems go away?

 

[3v0]

There are several programs on can build that are more dependable and not just for the voltage problem. The best of the lot are the ICD2 and PICK2 clones that use a uC to do the timing. Another big plus is that microchip updates the software to include new chips a they are released. They also work with MPLAB, have ICD etc etc etc...

 

[Sceadwian]

The problem with the ICD2 clones is they required a programmed pic... If you can program the pic to make the controller in the first place, where's the problem?

 

[3v0]

And that is why I said the only good use I can see for a JDM is programing the chip for a good programmer.



EDIT: And the first thing you should do after getting your first ICD2 or PICkit2 working is make another copy of the chip.

 

[blueroomelectronics]

I programmed my first Inchworm with an ancient and very slow PIC Start+

 

[Nigel Goodwin]

Right Nigel, as I said, if you supply the target board with power directly.. Don't the the problems go away?

It helps, there are many such modified JDM type designs about.

 

[Tightwad]

what adjustments can I make to the code to avoid the problem, while I wait for a new programmer? I would like to work through the tutorial if that is possible, and retain the ability to re-program.

 

[Nigel Goodwin]

what adjustments can I make to the code to avoid the problem, while I wait for a new programmer? I would like to work through the tutorial if that is possible, and retain the ability to re-program.

You could add an external crystal and capacitors, and use the XT setting in the config fuses. Also enable MCLR as a reset pin in the config as well.

But you've then lost three I/O pins, which is lowering it's usefulness.

 

[Tightwad]

Here is my end goal, maybe the correct chip to use could be suggested? I think the 16F88 has what I need, but then I am not sure. End result would be a chip that can take these inputs:

Inputs:
Switch 1, cycles fan mode (3 settings)
Voltage input for Voltmeter and Headlight override comparison
02 sensor bank 1
02 sensor bank 2
Throttle Position sensor
Brake Light

And generate these outputs:

Output pins:
Fan setting (grounded or high)
Modulated Brake Light output
O2 Bank 1 output (not sure how many will be needed)
O2 Bank 2 Output (not sure how many will be needed)
Volt Meter Output (not sure how many will be needed)
Headlight override output (pin high to switch relay)
Throttle Position Switch out (not sure how many will be needed)


Such that indicators can be used to display Voltage and Throttle Position %, Modulate the Brake (other circuitry needed of course), Display fan setting (OFF, ON, DEFAULT) mode, Display the Air/Fuel mixture (from the O2 reading) for bank 1 or 2.

Now I realize this is VERY ambitious, especially given what I know, but the plan is to gain an understanding of HOW to use the PIC, then work on each of these parts individually, THEN combine them.

I am not sure of course what other hardware, or how many output pins will be required to facilitate an LCD or LED display...lots to learn!

 

[blueroomelectronics]

What are you trying to build? Why are you playing with the O2 sensor?

 

[nickelflippr]

Here is my end goal, maybe the correct chip to use could be suggested? I think the 16F88 has what I need, but then I am not sure. End result would be a chip that can take these inputs:
...
...
...
Now I realize this is VERY ambitious, especially given what I know, but the plan is to gain an understanding of HOW to use the PIC, then work on each of these parts individually, THEN combine them.

I am not sure of course what other hardware, or how many output pins will be required to facilitate an LCD or LED display...lots to learn!

When incorporating an LCD display, and or LED's you will easily be over the pin budget of the 16f88. With an LCD in 4 bit mode, 6-7 pins are required alone. When completing the individual tasks at hand, a much larger device will become apparent. An 18fxxxx device and a high level language would be my choice to run the show, but that is a whole topic unto itself that gets discussed frequently on this forum.

My suggestion would be to buy the Pickit 2 programmer kit, which includes the low pin count board. To run through Nigel's tutorials, solder a 18 pin DIP socket to the proto area of the LPC, and hookup the Vdd, Vss, 0.1uf bypass cap, and ICSP connections. Add female 0.100" headers for on board analog pot/led's, and off board (i.e. breadboard) connections. Insert 18 pin device of choice and you are ready to go.