Getting a Thing-O-Matic going

[Jim]

Hi all

I picked up a Thing-O-Matic recently and have been going through the process of updating firmware to run ReplicatorG 0040.  I know these are old bots, but I'm determined to get this going as simply as possible.

Here are some specs for my bot:

MakerBot Thing-O-Matic 

Arduino Mega 2560

Stepstruder MK6

Extruder controller v3.6 (ATMEGA168)

Motherboard v2.4

Running ReplicatorG 0040

By the way - before trying any of these upgrades, I have done some basic calibration with the bot using an earlier version of RepG just to make sure it works and see what it could do. Everything is fine - XYZ, stage heating, extruder heating. However, i always got the warning that the firmware version on the EC and motherboard should be v3 or later, otherwise "proceed with caution"...

Upgrading has gone pretty smoothly and the final thing I need to do is update the firmware on my Extruder controller.  Problem is I continually get the "Upload not successful ..." error.  There is a lot of troubleshooting in groups across the net, but no definitive answer. [for example: https://www.reddit.com/r/3Dprinting/comments/2ntf10/just_received_a_makerbot_thingomatic_with_a/cmgs6ha/]

In updating my motherboard and EC, here is the sequence I've followed:

Running RepG 0040

1. Plug in motherboard

2. Scan for serial ports

3. Find the one the motherboard is using (COM4)

4. Update the firmware successfully

5. Unplug the motherboard from the usb cable

6. Plug the EC into that cable

7. Re-scan for serial ports

8. Find the one the EC is using (COM5)

9. Try to upgrade the firmware, to no effect

I am directly connected to the EC board. I have tried unplugging everything else from the EC board so that it is completely stand-alone, I have tried pressing reset on the EC board (even though that is not supposed to be necessary, some have reported it works for them) and various points before and after on the ATMEGA board. Nothing else is connected to my PC and the ATMEGA board is the only USB connection.

From what I've gathered reading other posts, there are a few possibilities:

1) there is no bootloader on the ATMEGA chip

2) somehow the communication between PC and board is not working properly (everything in PC control panel shows it is fine though ...)

I'd love to hear some possible solutions if anyone has any. 

Cheers, Jim

[MacGyver]

In the past flashing my TOM I've had to hit that reset button 7 or more times before it would start the update.  It eventually will start flashing rapidly and update.  

[Jim]

Thanks MacGyver,

I know the reset is required with the TOM motherboard, but not necessarily the Extruder controller. I just tried it in case, but no visible rapid flashing.  The little LED beside capacitor C14 is constantly flashing about once per second ...

[Jetguy]

That flash is just the heartbeat indicator of the now current loaded running firmware.

Flashing is simple if you understand what is happening.

The only way to write to flash memory is either a program running on the microprocessor (AKA bootloader) or via the ICSP ports.

You plug in the USB, that is a USB to serial conversion FTDI. The 168 or 328P (hint, hint, hint, hint- could be your problem as wrong processor is selected) and you ONLY communicate over serial to the processor. You cannot write to flash (there is no physical connection, just a logical one through the processor when the bootloader runs.

Bootloader ONLY runs for a few seconds after reset. The USB to serial can reset with a command, but there is also a manual override button that will reset thus making the bootloader run for a few seconds. You have a VERY small window of time where the software will try to connect to the bootloader and attempt to flash. However, if you have the WRONG flash selected- i's the wrong processor "type" sent to AVRdude flashing and you'll get the error because the hardware doesn't match the command.

SO, #1 KNOW what processor you board has (read the label on the processor NOT the silkscreen on the board!!!!!). Most came with a 168, some came with a 328 and they REQUIRE different selection to flash.

[Jim]

Hi Jetguy, this is definitely a 168 chip. Flashlight and magnifying glass confirmed. Though interesting that you mentioned the silk screen, which I just had a look at. It says 328.

I just tried a few iterations of pressing the update button on RegG and the hard-Reset button on-board with various intervals between pushing "update" and "reset" (from t=0 through to t=3s). No luck.

This is the update link I'm using: http://firmware.makerbot.com/firmware.xml which works fine for the motherboard update.

I tried selecting version 2.7 and 2.8 (rather than 3.1) but that also did not update.

Another bit of info: when I push Update, the heartbeat stops briefly and then starts up again (much like what happens if I push the on-board reset button)

What about the version of RepG?

Thanks, Jim

[Jetguy]

I don't think the version of Rep-G so much matters about this specific instance of flashing an extruder controller since all we are doing is using the XML from the firmware site and AVRdude which is not different between versions, but where you point at updates must be correct and i haven't checked, but maybe MakerNot messed up in the firmware repository online.

I'm one of the few people left that i know of with T-O-M and Cupcake electronics around but being honest, haven't touched this process for a while.
I know for a fact I could flash it if I had it in front of me, even if I had to extract the hex off of one of mine and flash using a programmer.
I know when I get one that's a pain, I just break out the programmer, flash the bootloader, and then flash the firmware and it all works.

I just dug out of the box a 328P one (version 3.6) and flashed using Sailfish Rep-G version 33 by changing the update URL to the makerbot link http://firmware.makerbot.com/firmware.xml
No fuss, it flashed easy as can be.

[Jim]

It's frustrating. I just can't flash it. I don't have a programmer, but will find a way eventually.

Thanks for the help!

[Daniel Newman]

On 04/01/2017 1:12 PM, Jim wrote:
> It's frustrating. I just can't flash it. I don't have a programmer, but will find a way eventually.
> Thanks for the help!

If you have 3.0 on the board you don't need 3.1. V3.0 will work just fine with MBI's final 3.1
motherboard firmware as well as the Jetty and Sailfish firmwares. (There's no reason to install the Jetty
Firmware since it's just an older version of Sailfish: the successor to 3.5 was renamed to Sailfish and
came out as Sailfish 4.0. V4.7 is the current version of Sailfish for Cupcakes and Thing-o-Matics.)

FWIW, I too find flashing the extruder firmware VERY difficult. 'tis why I've never bothered to try
to distribute the couple of firmware fixes I've made to the 3.1 extruder firmware: it would just be
inviting a lot of people to do something likely unnecessary for them and certainly frustrating to attempt.

Dan

[James Rabeau]

I would do that Dan, but the version on my EC is 2.6. None of the installs work for any of the versions either.

What problems do you encounter updating the firmware on EC? And how do you solve? Seems like it's a matter of luck with the reset button!

J

[Daniel Newman]

On 04/01/2017 1:50 PM, James Rabeau wrote:
> I would do that Dan, but the version on my EC is 2.6. None of the installs work for any of the versions either.
>
> What problems do you encounter updating the firmware on EC? And how do you solve?

I use an AVR ISP programmer when I must. Flashing the EC from RepG is to hit-or-MISS (mostly "miss").

Dan

[James Rabeau]

Got it!

[Ken Kirby]

All - I've recently dusted off my old Thing-O-Matic as well and also faced some issues upgrading firmware to Sailfish. I was able to update my extruder controller but was not initially able to update the motherboard MEGA2560.

After reading through the following links I was able to get it updated via command line in Windows7 in the same manner that neomantra did under OSX.

https://gist.github.com/neomantra/8413042

https://learn.sparkfun.com/tutorials/pocket-avr-programmer-hookup-guide/using-avrdude#device

http://www.nongnu.org/avrdude/user-manual/avrdude_4.html#Option-Descriptions

https://batchloaf.wordpress.com/2013/02/12/simple-trick-for-sending-characters-to-a-serial-port-in-windows/

This was the basic steps I took:

1) disconnect the extruder controller from the motherboard - disconnected the short blue RJ-45 connector. Note that I'm not sure this was necessary.

2) changed settings for COM12, which is where my USB connection was mapped, to use a higher baud rate of 115200 - this also may not be necessary.

3) in the console navigate to the directory under replicatorg that contains avrdude - for me that was "C:\temp\Sailfish\replicatorg-0040r34-Sailfish\tools"

4) in the console enter the following command to validate you can connect - if you enter this without first pushing reset, you get a timeout issue. I practiced pushing reset then immediately hitting enter. 
     avrdude -c stk500v2 -P \\.\com12 -pm2560
    (make sure you are not connected with ReplicatorG at this point or you may get an Access Denied message)

5) once you are comfortable with the timing you can run the following
     avrdude -c stk500v2 -P \\.\com12 -pm2560 -D -Uflash:w:"C:\Users\<your_user_here>\.replicatorg\firmware\MB-mb24-2560-Sailfish-v4.7-r1428.hex":i

Note that if you are connected via a lower com port number, you may not need the "\\.\" options...

Even though you still have to execute the command right after hitting reset, I think this approach is much less finicky than executing this update from within ReplicatorG - I did it on the first try. Hopefully this may help the next guy with this issue.

Ken