Fluke 8840A + Arduino?!?!

[int2str]

Another question about my "new" Fluke 8840A:
My meter does not have the GPIB option board - neither is that very useful for me. However, looking at the user manual, it looks like the GPIB board is simply communicating with the main MCU with an isolated 2-wire protocol, apparently already at 5V. So, seems like it would be very easy to take an Arduino Nano clone or something and hook it up to the meter to get USB/serial output and control instead of GPIB.

Has anybody done anything like this before? Any documentation on the 2-wire protocol from the expansion board to the internal MCU?

Thx 

[Brumby]

I don't know of any specifically - but Google is your friend.

I just plugged in "Arduino" and "GPIB" to get some promising returns in the search...

[int2str]

Just to clarify, I do not want the Arduino to talk GPIB at any point (which is what most search results are for).
I want to hook the Arduino up directly to the internal MCU on the 8840 and use it *instead* of GPIB.

The pins on the internal chip are labelled "UART". Just wondering if I can directly interface it with an arduino.
Suppose the hardest part would be to reverse engineer baud rate, protocol etc...

[bitseeker]

That sounds like a cool project. Sounds like you've got some scope/logic analyzer work ahead of you. Be sure to verify those voltages first.

[Brumby]

Ah, OK.

It does indeed sound like an interesting project.  Decoding the signals will be fun.

[jesuscf]

That sounds interesting.  Perhaps, since you don't have the GPIB option, it would be easier to disassemble the code of the Z8611 (U202) and see if something can be found there.   After googling around a bit I found the image for U222 but I couldn't see anything obvious.   Also it doesn't help much that I am not proficient in Z8 assembly.  So far no luck finding the memory image of U202, which from the schematic seems to be part of the micro-controller, and maybe code protected.

Another question about my "new" Fluke 8840A:
My meter does not have the GPIB option board - neither is that very useful for me. However, looking at the user manual, it looks like the GPIB board is simply communicating with the main MCU with an isolated 2-wire protocol, apparently already at 5V. So, seems like it would be very easy to take an Arduino Nano clone or something and hook it up to the meter to get USB/serial output and control instead of GPIB.

Has anybody done anything like this before? Any documentation on the 2-wire protocol from the expansion board to the internal MCU?

Thx 

[vbalint]

It's really an interesting project. I have also a 8840A with GPIB and TRMS options. Unfortunately I have only PCI GPIB card and I use a laptop usually. It would be nice if I could use my laptop with this puppy. Of course I can build a USB-GPIB thingy but this is also interesting. Maybe I can help with some information how the communication is going.
First of all the Instruction manual contains the schematic of the option board. The funny thing is that I have an another revision of the board with different ICs and layout but at least the connector pinout is the same You can use this schematic to create your own differential transmitter/receiver.

I'll try to sniff around with a TTL-RS232 converter in the protocol in the coming days weeks. I hope it is a human readable format. If it is I can find out the baud rate by trying. Or if someone has the option board and a digital scope please catch a transmission to measure the baud rate.

Arduino may not be the best option. The ATmega328P has only one USART and it is used for the USB communication too. From software point of view it is easier to use a micro with two UARTs than write a software UART.

I've attached the schematic and two picture from my board. Maybe they can help somehow. WARNING! It's not the same as on the schematic!

[int2str]

I think Arduino is a good choice because it has USB on board and can be had dirt cheap these days. I would lean more towards an Atmega32u4 here since it has USB built-in and the UART available. So for ~$3-5 you can get a clone.

Interfacing with the host MCU could be done two ways:
1. Directly to the UART, bypassing the "crossing guard" circuit. This could be done simply by hard wiring it in and would not require any circuit board. But of course would not be very modular. I will probably do this for debugging/development.
2. Custom PCB with crossing-guard section. If I can figure out the protocol, I'll probably spin a PCB that would allow remote logging, triggering etc.

[vbalint]

I've captured a data packet on the TP901 which is the TX side of the instrument. I've done it with Pickit2. See the attachment.
The DMM sends this packet approximately in every half second. It looks like it uses 7 bit transmission with 62500 baud rate. Correct me if I'm wrong.

I didn't know there's Arduino with Atmega32u4. That will be fine

Unfortunately I don't have more time so that's it for today.

[vbalint]

Unfortunately the Pickit2 UART tool's maximum baud rate is 38400

[Andreas]

Hello,

I see 1 Start Bit (low) 16 us
8 data bits (last is low) until cursor 2
the Stop Bit(s) are after the cursor 2 as high level.

But caution the first stop bit could also be the "adress marker" of a 9 bit protocol,
if it is the first byte of a sequence. Especially when a 8051 CPU is on the main board.

With best regards

Andreas

[int2str]

Poked around my meter a bit as well.
So yes, it looks like the baud rate is 62500.

According to the Zylog manual:
"The bit rate is controlled by Counter/Timer 0, with a maximum rate of 62.5K bits/second.
The Z8611 automatically adds a start bit and two stop bits to transmitted data (Figure .
Eight data bits are always transmitted"

[softfoot]

Did this project come to anything in the end ??

Looks very interesting :-)
Dave