Disabling Axis

[rojhan]

The fundamental problem appears to be with the C11G.

With everything disconnect from C11G other than power and the LPT connector from the UC400ETH:
- Charge pump disabled. C11G commands outputs to GND and +5V
- Charge pump enabled. C11G commands outputs to GND and +1.7V

[rojhan]

What is the version of UCCNC and which screenset do you use?

@deszoe I'm actually chasing two different problems, and one of them does seem to be related to UCCNC.

With everything disconnected, other than power and motion controller, as described in a previous post, UCCNC/C11G is setting the enable pin to low/0V/ground when I untick the "axis enabled" box for the A-axis, IF the charge pump is enabled. The other enabled-enable pin for the XYZ access behaves as expected/desired.

I'm trying to find a way to look at the PC-side of the LPT port to see what the actual output is there.

[dezsoe]

I asked the version and screenset to check if there's any bug in the screenset, but it's OK. Now could you upload your profile? Maybe, I can find something in it.

I use a UC400 daily and never had any output level problems. I don't have a C11G, but I'd check your profile using a UC400.

[rojhan]

I asked the version and screenset to check if there's any bug in the screenset, but it's OK. Now could you upload your profile? Maybe, I can find something in it.

I use a UC400 daily and never had any output level problems. I don't have a C11G, but I'd check your profile using a UC400.

I've started a support discussion with CNC4PC. My initial, VERY preliminary, thought based on what I've received so far is that the 12Khz charge pump signal is bleeding through to the enable output pin when UCCNC disables the axis.

Profile attached for review, though.

Thanks!

[rojhan]

I simplified the troubleshooting as much as possible. Everything below the UC400ETH has been disconnected and I went directly to the IDC26 connector on the UC400ETH.

- If the Axis Enabled checkbox is checked, the output pin state follows the active high/low checkbox and reset button status - Good!
- If the Axis Enabled checkbox is unchecked, the output pin state is low, regardless of the reset button status and the active high/low checkbox status. - Not really good...

[ger21]

What I have found is that enable/disable on the A Axis (XYZ as well) does not actually follow the configuration of the enable pin active high/low. If the RESET button is not active (not reset) the stepper remains enabled even if the axis is disabled regardless of the configuration of the high/low configuration of the enable pin.

Did you click Apply or Save Settings after unchecking the box? Nothing happens when you check or uncheck a checkbox until you Apply or Save the settings.

[rojhan]

Did you click Apply or Save Settings after unchecking the box? Nothing happens when you check or uncheck a checkbox until you Apply or Save the settings.

Yes

[dezsoe]

It took me a long time, but finally I understand your problem. We have to discuss it, because both you and UCCNC are right. You're right: when you disable the axis then you need the inverted output to turn off the driver. UCCNC is right: when an axis is disabled then the outputs and inputs for that axis are ignored.

[rojhan]

It took me a long time, but finally I understand your problem. We have to discuss it, because both you and UCCNC are right. You're right: when you disable the axis then you need the inverted output to turn off the driver. UCCNC is right: when an axis is disabled then the outputs and inputs for that axis are ignored.

Working backward from the troubleshooting, everything is working as designed, just not necessarily as desired or in cooperation witch each other.

- With no control input input, the default output pin states on the UC400ETH is low.
- The stepper drivers need a relative high on the ENA+ pin to *disable* the outputs
- With the C11G *not* in charge pump mode, its outputs follow the UC400ETH state; default low
- When the C11G *is* in charge pump mode, and not receiving a 12Khz signal the output pins are disabled and float high.

With all of the above, I originally connected ENA+ to 5V and ENA- to the output pin. With the default low, until commanded high by UCCNC (enable pin active high), that kept the steppers disabled. There is still, however, a brief period when UCCNC starts up that the output pin blips high and the steppers thunk.

With the C11G charge pump mode turned on, the behavior of the output pins changed, but still as spec'd. Without the 12Khz signal coming from UCCNC, the C11G disabled the outputs. The disabled outputs are high. That meant that my previous stepper connection enabled the driver (high from the disabled output and high from the +5V) which, because there was no relative difference, enabled the driver.

To "fix" the problem, I moved ENA- to ground and ENA+ to the output pin (the pullup resistor was part of troubleshooting and, ultimately, a red herring). Without a charge pump signal, the stepper drivers were disabled because of ground on ENA- and the floating/high on ENA+. I changed UCCNC axis to enable pin active low. Until UCCNC is in control, sending the charge pump signal, and pulling the axis enable low, the stepper drivers are disabled.

This finally led to current state where I disable the axis enable checkbox and UCCNC does the right thing and doesn't touch the output. The UC400ETH does it's normal thing and an idle pin is low. C11G does the right thing and follows the UC400ETH state. Stepper drivers do the right thing because they are receiving a low/low and enabling the stepper. Which isn't what's wanted.

I would like to see an additional option in UCCNC to allow ignoring the enable pin or forcing it to the inverse of the enable pin active level. Keep the existing behavior as default to not break any existing installs.

Now that I better understand what is happening, I'm going reconnect ENA+ to 5V and ENA- to the output pin and pull the output pin low with a resistor. With no inputs this should disable the drivers (ENA+ at 5V, ENA- pulled to ground). Change UCCNC back to enable active high. See if the C11G will overcome the pulldown and activate the steppers. That's an exercise for tomorrow, though.





This is

[fsli]

I'm going reconnect ENA+ to 5V and ENA- to the output pin and pull the output pin low with a resistor. With no inputs this should disable the drivers (ENA+ at 5V, ENA- pulled to ground).

Here's what the DM556T manual says:

There appears to be a contradiction between what you're expecting (ENA+ disables the driver), and what the driver manual indicates (ENA+ enables the driver). To clarify how it operates, I would completely disconnect the driver from the C11G, then manually put +5v on the DM556T ENA+ terminal, and ground on the ENA- terminal. If the stepper engages then you know +5v on ENA+ is an enabling, not disabling, condition. Then force both ENA+ and ENA- to ground, and that should disable the driver. Remove all connections from ENA+ and ENA- and that should enable the driver again.

I can see the confusion in how the DM556T manual is written. It mentions PNP signals inverting the logic, but the example schematics all show +5v on ENA+. The problem is controlling the startup condition so that the DM556T doesn't see a disconnected state on ENA+/-. The solution is to leave ENA- connected directly to ground, and use one of the C11G relays to drive the ENA+ terminal. Connect the relay common to ENA+, connect the relay NC terminal to ground, and the relay NO terminal to +5v. Then configure the axis Enable pin/port to whichever pin/port drives the relay.

In that configuration, both ENA+ and ENA- will be connected to ground at startup, and the DM556T should disengage the stepper. It should only engage the stepper when the relay control pin is asserted, and only if the charge pump is running.

P.S. -- Stop pulling pins to +5v or ground. You're likely to create more problems that could blow up your driver or C11G.