EB-006-00-9 Driver issue

[Crenwick]

Hi all,

Well I just received my EB006-V9 and I have the same problem (when programming a chip Error "1: Flash memory not verified - may be due to code protect").
I go to mloader to check directly the board and USB driver and tested the hex file wich produce the same error :

V9.jpg (79.06 KiB) Viewed 15303 times

So I took back my EB006-V7 board and did the same test and .... It WORKS !!!

V7.jpg (71.54 KiB) Viewed 15303 times

Erasing device seems to works on both but writing go wrong. Also the chip doesn't start, so I imagine that the written data is garbled.

Should be interesting to log the USB communication between mloader and EB006 to see if the problem is mloader or EB006-V9 ?
I will try to find an USB logger on the net. If someone has an idee or want to help me no problem

[medelec35]

Hi Bernard,
Have you seen This
It solved my EB006 V9 issues

I don't believe drivers for V7 board work V9 board and vice-versa
So that sounds like you may have not have correct driver installed?

With V9 you need to download the correct driver.

Drivers can be found here

Martin

[Crenwick]

Hi Martin,

Yes I saw the post

But I forgot to mention that I now use only FC6 so I didn't try to install the V9 board on FC5.

I also uninstalled the two USB drivers and reinstalled them afterwards. The cards are beying recognized (one at a time) without problem but like you can see above I directly used mloader.exe to test writing an Hex directly.

When you installed the V9 board with FC6 did you had these problems ? If yes, can you explain how you solve the problem ?

Regards,

[medelec35]

Hi Bernard,
Using mLoader from Q&A Link
as a stand alone programmer for directly loading hex files there have been no issues for me:

mloader from Q&A.png

(96.19 KiB) Downloaded 8880 times

I will try with other chips and let you knoe how I got on.

As a test run, have you tried with the 16F1937 that came with the programmer?

if that works then as a guess, perhaps mLoader can do with additional Vdd switching options that PPP had?

Martin

[Benj]

Hello,

Yes the verification warning is on a slightly earlier version of mLoader, the latest mLoader is now available to download from our resource centre and the EB006 product page as well as being packaged with Flowcode 6.0.7.

if that works then as a guess, perhaps mLoader can do with additional Vdd switching options that PPP had?

Yes we do need to add this, I'll check it's on the list.

[Crenwick]

Hi Martin & Ben,

Yes I'm using the mloader from the same link and the latest USB driver with pdf and doc files info included in zip.

I also tried to use other pic's like 16f628 and the 16f1937 that comes with the board, all ways same message

f1937.jpg (89.38 KiB) Viewed 15279 times

I tried also on my laptop, no success.

The last thing I did now is to use another USB port on my desktop and it worked 1 time until I remove the USB cable and plugged it back in and the error came back !

I tried two other ports with no success. Changed also the USB cable (why not).

I don't know what to do further... Ok, I have my old V7 board that still works but it's not a solution.

Very very strange !!!

[Crenwick]

Problem found !

What I did is :

1. plug in my V7 board
2. Goto device manager and uninstalled the Pic Micro Programmer device with remove of his drivers !!!
3. Plug out the V7 board
4. Plug in the V9 board

And tadaaa, board is WORKING

I also did the same thing on my laptop (where I had also FC5 and V7 board installed) and it works also !

So now to the Staff, did you tested the V9 board on a system with the two USB drivers installed ?

And now time to work to my project

[Benj]

Hello,

I have both USB drivers installed on my test PC here and working well but I only plug one EB006 board in at a time.

Could be that mLoader is getting confused about which board to send the program to if you have more than one board connected at once?

[Crenwick]

Hello Ben,

No, I never connected the two boards at the same time.
Tomorrow I will try to re-install the USB Driver for the V7 board on my laptop to reproduce the error.

I'm not sure but mloader seems to detect the correct board but when he starts the communication with the USB interface it gone wrong.

I saw that the PID from the boards are different 020 vs 021 and VID are the same (logic) but maybe it's a tip to look in mloader if ...

Regards,

[bobingabout]

I'm not at work for the next couple of weeks, so I can't test anything right now, but all these later tests (with debug info posted at the bottom of page 1 of this topic) were done on a fresh copy of windows 7 (as installed by a guy from IT services), I then personally installed MPLab V8.92, mLoader, and the V9 drivers... nothing else. and those were the problems I was getting.

[medelec35]

I have got a Eb006 V7 and EB006 V9 drivers installed on the same pc (64bit win7)
This did not affect programming V9 via stand alone mLoader software.
Both boards are detected correctly.
Only time V9 failed was when both programming boards was plugged in at the same time.
That was expected since they are not designed to be plugged in at the same.

[Crenwick]

Good Morning !

This morning as test purpose I did the following :

1. Test my V9 board --- ok
2. Install again the V7 driver and retest V9 with mloader -- FAILED
3. Uninstall V7 driver and test V9 board -- FAILED but with a "Unknown chip type in datafile"
4. Uninstall and reinstall V9 driver with power off/on of board --- WORKS AGAIN !

I did this two times with same result

What I also saw is in the device names that there are two names
USB\VID_12BF&PID_0021&REV_0000 and USB\VID_12BF&PID_0021

mloader is adressing USB\VID_12BF&PID_0021&REV_0000

Maybe a good idee to look in this direction ? Ben ?

Regards

[Crenwick]

Bad news !

I had to restart my desktop for windows updates and now my V9 doesn't work anymore ...

It's really disappointing , I tried five times to uninstall/install the driver without success, even on my laptop it stopped working.

More and more I think that the on-board PIC24 is doing strange things on the USB.

I now just tried on 4 computers (win7 32 & 64) and all the same errors.

Err2.jpg (132.62 KiB) Viewed 15241 times

@Matrix , do you have a tool to update the on-board PIC24, maybe bad programmed ??? Help please !

[Crenwick]

Hi all,

Worked some hours this weekend without success on V9 board

Who is working at Matrix on Mloader ?

Is Mloader and V9 board working without any problems on your systems ?

Regards,

[Crenwick]

Hello,

I tried something today, I programmed a 16F1937 with a simple 500ms blink led on RB0 with my EB006-V7 board. Then I switched from board to the V9 and tested the ICT function and it works !

ICT.jpg (92.36 KiB) Viewed 15205 times

So I suppose my V9 board should be ok ? I'm more and more convinced that the problem comes from the mloader but why is it working by others like Martin and Matrix staff ???

If I can help you with other tests let me know

[kersing]

Hi Bernard,

I has issues with mLoader as well, after uninstalling the drivers for LEAP Motion all issues disappeared. When looking at the Windows device manager, are there any Human Interface devices listed?

Jac

[Crenwick]

Hi Jac,

Well like you can see I have a lot of HID interfaces.

HID.jpg (80.84 KiB) Viewed 15197 times

I'm only using one mouse and one keyboard from Logitech .

What do you think ?

Regards,

[medelec35]

Hi Bernard,
I don't want to mislead you.
There are problems with mLoader as Jac as stated.
I believe matrix are probably working on them?
Still it would be nice to know the official matrix line?

When I have had an issue with a warning message about bad response popping up, I have to remove and reconnect power & USB then program does get sent successfully.

Martin

[petesmart]

Hi Bernard,

Hi Bernard,
I don't want to mislead you.
There are problems with mLoader as Jac as stated.
I believe matrix are probably working on them?
Still it would be nice to know the official matrix line?

When I have had an issue with a warning message about bad response popping up, I have to remove and reconnect power & USB then program does get sent successfully.

Martin

I have had a similar experience to Martin... Power off/on seems to be to interim answer.

ICD stimulates the problem further....

Best.

Pete

[Crenwick]

Hi all,

Thanks for the feedback, I will try to contact the staff.

[Benj]

Hello,

I has issues with mLoader as well, after uninstalling the drivers for LEAP Motion all issues disappeared.

LEAP motion hardware now on order so we can investigate why this is causing the EB006 v9 communications to fail when installed. Looks like it might also be a nice interface to Flowcode so bit of a double barrelled reason to buy it.

We are aware of a problem after using ICT/ICD where mLoader is unable to connect to the EB006 v9 probably because the handle for the USB connection is not being released correctly. We are looking into resolving this.

We are also aware of an issue where sometimes Windows can get there first with it's own driver which then seems to be a bit troublesome to get rid of. We are using our own USB IDs so not sure why Windows is jumping in with it's own driver but I have noticed that Microchip have recently and silently dropped the WinUSB devices from their examples portfolio.

[Crenwick]

Hi Ben,

Nice to have some feedback

Do you plan to use an alternative to winusb ?

Regards,

[Benj]

Hello,

WinUSB is a Windows conversion from the linux based LibUSB and is meant to be one of the ways forward, allowing fairly high bandwidth communications on USB 1.1 and USB 2.0 FS connections.

From what I have found so far it looks like the fact that Microsoft is installing it's own driver is a "new feature" of Windows 8 which has subsequently been rolled out to older systems right back to XP via Windows Update. The idea is that you don't have to manually specify the .inf driver file which in an ideal world would be fantastic.

Our software i.e. Flowcode Ghost and mLoader should both still work with the hardware with our driver installed or not so it looks like maybe we can tweak the software for mLoader and Flowcode at this end so it will work with either setup. I will do some more digging as soon as I am able and see if there is a quick win to allow us to proceed without having to go to the extreme of switching the USB data transfer class. It might be that as a fall back measure we have to go back to the older MCHPFSUSB data class which we have been using for years on our USB products (USB Slave component in Flowcode) but for now we will stick with WinUSB and see if we can find a simple fix.

The board is working great here on my Windows 7 64-bit setup and in fact on every machine in the building but I have reports that on some systems the Windows Update driver is overpowering our driver every time the computer is connected back to the internet. If you have the EB006 v9 hardware then I would love to hear your findings working with the product.

[Crenwick]

Hello,


The board is working great here on my Windows 7 64-bit setup and in fact on every machine in the building but I have reports that on some systems the Windows Update driver is overpowering our driver every time the computer is connected back to the internet. If you have the EB006 v9 hardware then I would love to hear your findings working with the product.

That should explain why it first worked and after a reboot no more...

Yes, I have now the EB006 V7 and V9 in stock I bought the V9 for the ICT functions and will sure test it in my next projects.

Until solution I will first program the PIC witch my V7.

[Benj]

EB006 v9 device in device manager installed with the Matrix driver.

DriverOK.jpg (6.57 KiB) Viewed 15118 times

EB006 v9 device in device manager installed using the automatic WinUSB driver.

DriverMS.jpg (5.33 KiB) Viewed 15118 times

Testing on my Win7 64-bit machine here mLoader and Test mode are both working with either driver installed?! Maybe the problem is to do with something other than the device driver?

We have experienced some issues with USB hubs so it might be worth changing the USB port on your PC if you are experiencing any problems with the v9 board. I always go via a hub and it is working fine for me but a colleague is using the exact same model of USB hub and for some reason this isn't working with the test mode?!