Hi WT,WT said:
To be complete, all you need to flash the µC, is the onboard power supply (5V reg and caps), the oscillator circuitry (crystal and caps) and the reset circuitry (classical RC components), plus the connectors.
If your µC board is complete, you just need to provide the digital power supply. That can be, as said, a simple 9V battery.
In this case, all you can test is the detection of the µC with the JTAG interface and the good flashing control test.
With the LCD display , IR receiver and wake switch, you can test the main functions. For a complete test, you will also need the encoder.
The IR receiver is mandatory, as it is wire to a NMI input.
You can simulate the wake switch with a touch of a screwdriver.
Cheers,
Trung
I finished the U-controller board and power supply. All of the power supply work as they should be. 5V on the u-controller board is working fine, led is turn on.
However, when I connect the program board and try the ponyprog2000 with steps from Cheff;
on Security and Configuration Bits when I hit read button
the program try to read and error occur
"Device missing or unknow device (-24)"
Am I missing something? How should I check what went wrong.
I use google to translate the ponyprog2000 file from French to English, so I can understand it. It is not perfect. May be I miss some points in there. Please help.
However, when I connect the program board and try the ponyprog2000 with steps from Cheff;
on Security and Configuration Bits when I hit read button
the program try to read and error occur
"Device missing or unknow device (-24)"
Am I missing something? How should I check what went wrong.
I use google to translate the ponyprog2000 file from French to English, so I can understand it. It is not perfect. May be I miss some points in there. Please help.
Connect your programmer board to your parallel port, but not to the target board (the µC board)
Lauch pronyprog, your µC settings (AVRmicro and ATmega64) should be recalled.
Click on the "mechanics wrench" icon, just on the right of the printer icon.
The setup dialog opens, check your settings wrt the pdf file, and click the "probe" button.
What is displayed ?
Lauch pronyprog, your µC settings (AVRmicro and ATmega64) should be recalled.
Click on the "mechanics wrench" icon, just on the right of the printer icon.
The setup dialog opens, check your settings wrt the pdf file, and click the "probe" button.
What is displayed ?
Hot news :
Two of the french diyers had the same problem (unknown device).
It happened that it was a bad db25 connector, at least for one of them, to be confirmed for the other...
Is yours the right one ?
It must be a MALE DB25 connector on the programmer, and a FEMALE one on the computer. Check if the pin one of the connector matches the pin one on the programmer pcb and on the computer...
Two of the french diyers had the same problem (unknown device).
It happened that it was a bad db25 connector, at least for one of them, to be confirmed for the other...
Is yours the right one ?
It must be a MALE DB25 connector on the programmer, and a FEMALE one on the computer. Check if the pin one of the connector matches the pin one on the programmer pcb and on the computer...
I agree with Cheff, had the same issue with my initial tries...verify that you have good contact in the connection with the PC port,....hope your cable is not too long as well...and sometimes if you are using serial port this may not let you use parallell conection.
I had all these problems, but you know...if it does not work, revert to basics, it's easier to identify causes.
I had all these problems, but you know...if it does not work, revert to basics, it's easier to identify causes.
It's confirmed for the second guy.CheffDeGaar said:
Wrong DB25 connector.
FYI, on a PC, a female DB25 is the parallel port, a male DB25 is a serial RS-232 port, but this connector has been simplified to a male DB-9.
The programmer is designed to be directly plugged on the PC parallel port, no need to use a cable between the PC and le JTAG interface.
Cheers,
Trung
...it means the programmer is there...but what it does not mean is that the programmer is "talking" to the display board.
worst scenario is either burned ATmega64 or no power on display.
...but first see if connection cable to display is ok and 5V is comming out of the display regulator.
( Be careful not to burn the ATmega64 with the multimeter ( it uses 9V battery ? )....ATmega cannot take that!)
worst scenario is either burned ATmega64 or no power on display.
...but first see if connection cable to display is ok and 5V is comming out of the display regulator.
( Be careful not to burn the ATmega64 with the multimeter ( it uses 9V battery ? )....ATmega cannot take that!)
I follow the steps as in Cheff's ponyprog pdf.
I plug in my programmer board to the parallel port but not connect to u-controller board yet. (I have to use my dad PC, it is the only one that has parallel port)
Open the ponyprog
Select the AVRmicro, then select the ATmega64
I click on the wrench icon and set parallel port with AVR ISP I/O and select LPT1 then click OK.
I also try probe as Cheff suggest and it said test OK.
Then I go to program options and select Reload files, Erase, Write Program memory (Flash) and Write Data memory (Eprom)
I connect the cable from programmer board to the u-controller board and power the u-controller board up.
Then I click the lock icon (security and configuration bits) and hit Read button. Then error occor.
"Device missing or unknow device (-24)"
This is how I got so far.
Thanks so much for all of you that help me go through this. Otherwise it would be like shooting in the dark.
I don't remember if I check the ATmega64 with meter or not. There are a lot of soldering and checking. You know, so many board. I don't mind if the ATmega64 is damage, I just want to know the source of problem and fix it.
I plug in my programmer board to the parallel port but not connect to u-controller board yet. (I have to use my dad PC, it is the only one that has parallel port)
Open the ponyprog
Select the AVRmicro, then select the ATmega64
I click on the wrench icon and set parallel port with AVR ISP I/O and select LPT1 then click OK.
I also try probe as Cheff suggest and it said test OK.
Then I go to program options and select Reload files, Erase, Write Program memory (Flash) and Write Data memory (Eprom)
I connect the cable from programmer board to the u-controller board and power the u-controller board up.
Then I click the lock icon (security and configuration bits) and hit Read button. Then error occor.
"Device missing or unknow device (-24)"
This is how I got so far.
Thanks so much for all of you that help me go through this. Otherwise it would be like shooting in the dark.
I don't remember if I check the ATmega64 with meter or not. There are a lot of soldering and checking. You know, so many board. I don't mind if the ATmega64 is damage, I just want to know the source of problem and fix it.
WT said:
Hi again WT,
I do no see the calibration step, this should be done, ie: close aplications other than Ponyprog and choose : Setup, Calibration.
Apart from that remember to check the pinout of both ends of the cable connecting the display to the programmer ( might be flipped! ).
From what you say, i gather that you might not be aware of how delicate the ATmega is; soldering should be with care not to damage the chip with static from the soldering iron and care should be taken not to overheat it as well.
Another issue that is sometimes forgotten in the "heat" of trying to get the #$%%&&/ thing to work is that while it can be measured it cannot support more than 5v and most multimeters have 9v batteries.....this can and will damage the chip is certain circunstances.
It is clear that your problem is between programmer and ATmega...now we just have to determine where!
Try: calibration/read security bits.
If this does not work and you are sure that the cable is good ...i would start to think that your chip is blown.
..but i hope i'm wrong.
Just got a similar case yesterday (unknown device).
I was sure that the programmer was OK (It's the one I've been using for yonks)
The µC board seemed OK, but it seemed that the µC soldering was a bit weak : The µC has been soldered using the "manual wave soldering technique" : applying flux, positionning the µC, soldering two opposite pins to hold it in place, solder rapidly all the pins on one row (being generous with solder) and remove excessive amount of solder with braid to clear the bridges.
The problem was that this diyer has been very generous when sucking the excess of solder off the board with braid : There was almost no solder left on the joints, just a haze of solder.
I just brought a small additional amount of solder on each pin of the µC, one by one, and checked the solder joints with a magnifying glass.
And simply doing that, the programming was sucessful, while a preliminary visual inspection didn't show open circuits on the soldered pads...
If you've been using the same technique, just reflow your sodered pins, taking your time and letting the µC cool down between each pin soldering.
This was how it looked like before reflowing the joints :
Tell us how it goes.
I was sure that the programmer was OK (It's the one I've been using for yonks)
The µC board seemed OK, but it seemed that the µC soldering was a bit weak : The µC has been soldered using the "manual wave soldering technique" : applying flux, positionning the µC, soldering two opposite pins to hold it in place, solder rapidly all the pins on one row (being generous with solder) and remove excessive amount of solder with braid to clear the bridges.
The problem was that this diyer has been very generous when sucking the excess of solder off the board with braid : There was almost no solder left on the joints, just a haze of solder.
I just brought a small additional amount of solder on each pin of the µC, one by one, and checked the solder joints with a magnifying glass.
And simply doing that, the programming was sucessful
, while a preliminary visual inspection didn't show open circuits on the soldered pads...If you've been using the same technique, just reflow your sodered pins, taking your time and letting the µC cool down between each pin soldering.
This was how it looked like before reflowing the joints :
An externally hosted image should be here but it was not working when we last tested it.
Tell us how it goes.