M44 input question?

[ger21]

I have my UC300eth and M44 sitting on my desk with nothing plugged into them.
With the new firmware, this is what I'm seeing on Port 4. Is this the correct behavior?

[beefy]

Good question Gerry.

I play with microcontrollers / electronics a little, and sometimes when inputs are not tied to either high or low levels, they can be indeterminate, or they go to one or another. Some microcontrollers can have weal internal pullup resistors activated too. Maybe breakout boards pull all the input pins to one level or the other.

Keith.

[ger21]

Keith,
The reason I'm asking is because there was an issue with the UC300ETH and the M44 motherboard, which was supposedly fixed with a firmware update in the UC300ETH.
Older versions of UCCNC did not have all of the Port 4 LED's lit like they are now.
Just want to make sure it's now working correctly.

[cncdrive]

Hi Gerry,

My understanding on the M44 is that all inputs are low when nothing is connected to them,
because there is an external pulldown resistor on the board on the inputs which is one resistor of a voltage divider.
The UC300ETH in the newest firmware has both the internal pullups and pulldowns disabled, the pins are floating, so only the M44 is pulling the pins.
It looks like on ports 1. and 2. and 3. it is like that, all inputs are low, but not on the 4. port.
I'm not sure if it is because it has a different inputs circuitry or what is the reason for this.

Unfortunately we do not have an M44 board. How we developing the firmware for the M44 is based on CNC4PC's description and discussion and at the final stage they test the board and give us feedback.
What we can do here is we can emulate their board using the UC300ETH module, identifying it as if an M44 was connected.

What I can do is I can ask Arturo at CNC4PC if this input LEDs state on port#4 is OK or not.

[ger21]

Ask Arturo to send you a board (Both the M44 and M44B).

It's important for you, because if the boards do not work properly, it's bad for CNC Drive.
It's also bad for Arturo, as I'm already a bit unhappy about my M44 purchase, especially now that I find out that it's never been tested by the developers.
While I realize that the fault of this lies mostly with Arturo, as he designed the board, but you don't want to be like Mach3 where you have a lot of compatibility issues.
You should either work more closely with Arturo, or tell your customers that 3rd party boards have not been tested by CNC Drive.

[Robertspark]

Unfortunately we do not have an M44 board. How we developing the firmware for the M44 is based on CNC4PC's description and discussion and at the final stage they test the board and give us feedback.

that's not good and must be time consuming for you with back and forth communication for the sake of postage and a few free issue development boards, could have got it right first time [or near enough] with the hardware in front of you.

[ger21]

I just sent Arturo an email, saying that it's unacceptable to be selling boards which have not been tested by CNC Drive for comaptibility.
Hopefully, he'll send you some boards quickly.

[cncdrive]

You know we also made the development the same way for the UC300USB and there was no problem.
We wrote the firmware and added the board to the API and CNC4PC tested it and confirmed that it was OK and we did not hear about any issues with that.

With the UC300ETH we did the same kind of development, but CNC4PC confirmed only with the latest firmware, probably they were too busy to test in the first few weeks of the UC300ETH lifetime, I mean it was a brand new board then and we were all busy with introducing the new board to the market.

And we did not think about the possible stronger pull resistors in the new microcontroller, I mean it pulls the pin with lower resistor value is why the original problem happened,
the same config worked fine with the UC300USB, so we did not think about this possible problem.

Then I discussed the issue and asked for some circuit diagrams from Arturo and then we found out what the problem is and the solution was to disable the internal pull resistors and leave that to the M44 pulldowns.

What I don't know is if the circuit differs on port#4, maybe it does and that could be a cause for the different non connected pins polarity, however this worked OK on the USB version, but to what polarity the pins are pulled on the M44 I don't know it either, so I'm a bit confused... will have to talk to Arturo about this to see it more clear.

And you are right that having an M44 here would make things much easier.

[ger21]

About two weeks ago, I plugged a cheap breakout board into the M44, and could not get the inputs to work. I thought it might be the breakout board.
I ordered a 5LPT motherboard, and when it came today, I plugged the same cheap breakout board in, and the inputs work fine.
So it appears that the inputs (At least on Port 1) do not work on the M44 motherboard.
I sent a similar message to Arturo.

[cncdrive]

Gerry, the M44 has external pulldowns on the inputs which is 2 resistors forming a voltage divider to divide the 5V input voltage for the UC300ETH module to 3.3Volts.
On the CNC4PC C62 BOB there is a pullup resistor, a much stronger one than the pulldowns on the M44, so CNC4PC's idea is to have the inputs low when nothing is plugged into the M44.
And pull the inputs high when the C62 is connected to the M44 and when the inputs are still not triggered.

If you connect a 3rd party breakout board to the M44 then the inputs are not working probably because the BOB has optocoupler inputs with standard NPN outputs of the opto and probably there is no pullup, because the designers of the BOB wanted to just leave the pulling up on the LPT port or the device which is connected to the BOB.
This is totally understandable logic, because without the pullups and if the output of the opto has a high voltage rating then the BOB's input optos' outputs could be used to work with not only with 5V, but even with 12V, 24V etc. But the missing pullups will make the M44 inputs not to work.
So, the M44 as far as I know was designed to be used with the C62 and other boards of CNC4PC and not for 3rd party BOBs or at least their goal was to design a board which they can easily fit to their own BOBs.

The -5LPT motherboard has a pullup resistor and a voltage divider, so all inputs have 3 resistors in total is why it is working even without the pullups on the BOB.
Our goal with the -5LPT board was to make an interface which will work with any BOBs which work with an LPT port.