E-Blocks PICmicro USB multiprogrammer V7

[medelec35]

Thanks for link. Doing nothing different after reading guide. Here are steps taken. Perhaps you would be able to see if doing any thing wrong.

Only have target chip on programmer, no other hardware connected. Not using RB7 or RB7 in flowcode.

1. psu and usb connected.
2 Jumper (ICD/USB) on USB since does not programming chip on with jumper on ICD.
3. Have target selected for chip....16F737 and ICD box ticked, pressed restore defaults.
4. Compiled to chip. Get a warning message about Hex file has config2 word, but selected chip does not. Hmmm must be that ppp bug , since 167737 does have two config words. After pressing continue, compiler window shoes programmed and verified OK.
At this point instructions does not say anything about putting Jumper (ICD/USB) on ICD. Logic suggested I should. But still made no difference! with jumper on USB or ICD When press step into, or run just got same message program on the chip is out out date.
Yet since chip was programed and verified, communication is OK.
Something obvious I am missing perhaps?

Was not picked up on that point so assumed is correct.

It did work with link in icd2. without link, is working 50/50 do think that will improve if
1)link stays in USB and not ICD2 and
2) speed is changed.
if 2, is correct speed trial and error? I did notice speed value does change automatically when different osc speed is selected

[medelec35]

YAY Now working OK at the moment 100% of the time, even on speed 42. Maybe speed issue was a red herring and it was jumper after all? I was changing it,when I should have left it. if so my bad. If something does not work, and I see the words ICD2, If instructions don't say leave in USB, then I thought best to put in ICD.
I know I was getting annoyed and that comes out in my text. Sorry about that.
Thanks for all your help.
I will let you know how its gone on Monday.

[Benj]

Hello

Sorry for the delay in replying our internet has been down since Friday afternoon and has only just been fixed.

Thats fantastic, great news its all fully working for you now. I can see how the silk screen on the board can be a bit confusing I will see if I can improve this for future boards. The ICD2 jumper is for the Microchip ICD2/PICkit hardware to connect allowing our USB controller chip to be bypassed. It is this USB controller chip that is also responsible for the ICD so moving the jumper would indeed sever the link to our ICD support.

Thanks for letting me know anyway I was worried about this one.

[medelec35]

At least it's sorted now. Thank you for your continuing support,it is appreciated.
What appears to be obvious to those who have already successfully used a product, when some one is using a product for the 1st time, this obvious thinking is lacking.
I have read about other products also laking in instructions.
In my opinion, the better instruction are, less support is required. How about good step by step guides that go with products. If a person follows a good step by step guide and product works first time, then it would be a great guide. Keeping Jumper where it is, keeping external psu connected and using component properties to set osccon, when using internal osc should be included. last one definitely not obvious, but all contributes to create a working product.

[medelec35]

ICD is really great, It has help me out a lot. Even without C, the simulation calculation results can be so different form the calculations performed within the hardware. Reason for this appears to be simulation rounds up and down, the hardware does not. The down side is with 18F4455, For some reason Timer0 interrupt does not fire and serial 232 (Via a different software pin) does not work when in ICD mode is selected, Even if Run is selected.
If I then un-tick ICD mode, and re-send to chip, Timer0 Int and 232 works again. Is there a way around this please?
Much appreciated.

[Steve]

Can you please provide a few examples for the ICD TMR0 and RS232 problems. And also I'd like to know more about the simulation problems - please provide an example of this too.

{edit} I've just noticed that you've given an example of the simulation in another post!

[medelec35]

Re: RS232.
Sorry It was not the fault of ICD, it was a hardware issue. Since using a different USB232, the issue has gone away.

Re: Timer0.
Timer0 is the only interrupt I am using. Not using any macro calls. When compile to chip without ICD enabled, there are no warnings of stack corruption.
When ICD is enabled I get the following message:

Building CASM file

Serious Warning: Possible sw stack corruption, function 'FC_MARKER' called by more than one asynchronous thread (main/Task, interrupt, interrupt low)

Serious Warning: Possible sw stack corruption, function 'ICD_PUSH' called by more than one asynchronous thread (main/Task, interrupt, interrupt low)

When I delete Timer0 interrupt and enable ICD the warning message disappears.
ICD appears to work ok with a short code running. When running larger code, timer0 fails to trigger after a short period of time
It could be something I have done wrong?
I have attached a very watered down version which will show the warning message.
Note only using 18F4455 at 19.66 Mhz, so can have a lower frequency PWM, otherwise would run at 48MHz

[Steve]

Ahhh - I see. Stack corruption could indeed be the issue. This will require a fix to the Flowcode exe itself, so I can't provide a quick fix for this.

[medelec35]

Thanks for looking into it anyway Steve.

[medelec35]

Sorry for bringing up the issue of corruption and ICD. I am using the MIAC for all different tasks. Flowkit works great if Timer0 not used, but unfortunately I do use it a lot, and all sorts of problems randomly occur due to corruption with the timer.
Is this fix anywhere near please?

[Steve]

We are currently finishing off the Flowcode for dsPIC/PIC24 product, so we don't have the time at the moment to solve this one. But we will look at it in the new year.

[medelec35]

Thanks for your reply.
If you required a beta tester,then i could help.

[Steve]

Thanks for the offer. In fact, I was going to send it to you anyway as a "thank you" for your help in answering the posts on the forum!