Dynomotion Forum

Hi Tom,

I'm having a problem with using Kanalog JP2 inputs.

I have connected 0V to the -ve side of the diff pairs and am connecting 5V signals from normal mechanical limit switches to the +ve side. A multimeter says that the voltages are as expected - normally 5V across the input and 0V when the switch is actuated.

The trouble is that on the KFLOP Dig I/O screen, Bits 36-43 stay high continually, no matter whether the switch is actuated (0V), not actuated (5V)or completely absent (presumably a solid 0V since the manual says there's an internal 470R across the input which would act as a pulldown in the absence of any connection?) The manual says that they fire up as inputs and I'm not aware that I'm doing anything to change that. The manual also gives me the impression that Kanalog just passes them through to KFLOP. Is there some configuration step I've missed which is required to make this happen?

I have Encoders which appear to be working happily on JP1 at 5V, so I presume that 5 is enough for these inputs.


Speaking of Encoders, I have the initial 'pulses to distance' figures measured, but can't for the life of me work out where to input the numbers to tell KFLOP how to read the Encoders. I'd be grateful if you would point me to the manual page (or whatever) I should have found this info on please.

Thanks very much,
Peter

Hi Peter,

The differential inputs must be driven with a positive and negative voltage with an absolute minimum of 200mv differential. That is the + input must be driven to higher than +200mv relative to the - input or to -200mv relative to the - input. A normal differential driver drives one input to ~ +0.4V and the other to ~ +3V and vice versa to obtain ~ +/- 2.6V differential. Consider using Kanalog's opto inputs.

The axis resolution would be entered into whatever application you will be using. For KMotionCNC they are entered into Tool Setup | Trajectory Planner | Axis Parameters | Counts/inch as described here.

Ok. Thanks Tom.

I had the Opto inputs earmarked for other operator panel type switches.

I'm now tossing up between providing pulldowns to a negative voltage on the JP2 inputs, and 'DACing' some of the other switch inputs onto a JP6 ADC input to free up some more Opto ins. I presume that ADC input could then be decoded back to individual switches (dig inputs) on the inside easily enough with a C program. Is that a reasonable assumption?

Maybe I don't really need the panel switches, but they are already there and there's something about a tactile switch you can easily hit with a finger instead of having to chase one on a screen with a mouse, especially in the heat of the moment.

Thanks again & have a great day!
Peter

Hi Peter,

I presume that ADC input could then be decoded back to individual switches (dig inputs) on the inside easily enough with a C program. Is that a reasonable assumption?

It should be possible to decode ADC values to create Virtual IO bits with C code running in a forever loop.

Consider adding a Konnect which would provide opto isolation and standard 24V signals.