KFlop + KStep integration

[John Coloccia]

I'm starting to integrate KFlop and KStep together...I thought I'd start a thread for the various questions that I'm bound to have...including this one .

I see that KFlop passes 12V from the power connector through to the KStep board, but the +12 seems to not be connected anywhere. Is that true? Does it go to a pad somewhere that can be used? It would be awfully convenient to drive the optos with that +12 instead of having to bust out another 12V line, into the breakout and back into the board.

Similar Threads:

• Newbie- Is it possible to use RS232 with Kflop and Kstep
• Newbie- KFLOP and KSTEP Three Axis Wiring
• Kflop+kstep out of syn
• Kflop+Kstep video
• General KFlop/KStep Precautions

[TomKerekes]

Hi John,

I'm not sure I fully understand your point. But no KFLOP does not require +12V. If +12V is applied to the KFLOP JR1 it is passed through to other connectors but KSTEP does not receive it.

It is best to use an isolated +12V supply to drive the opto inputs on KSTEP J33. That way all the limit switch wiring can't pickup and introduce noise into the KFLOP/KSTEP grounds or supplies.

Regards

[John Coloccia]

Hi John,

I'm not sure I fully understand your point. But no KFLOP does not require +12V. If +12V is applied to the KFLOP JR1 it is passed through to other connectors but KSTEP does not receive it.

It is best to use an isolated +12V supply to drive the opto inputs on KSTEP J33. That way all the limit switch wiring can't pickup and introduce noise into the KFLOP/KSTEP grounds or supplies.

Regards

Thanks for replying, Tom. My original thought was since KFlop will accept +5 and +12 through the Molex, and it passes that +12 through to J33, that it would have been convenient to power the optos from that +12. Are you suggesting that I shouldn't use a a PC power supply to drive both the KFlop/KStep and the optos but rather should use a 2nd power supply to isolate the returns?

[TomKerekes]

Hi John,

+12V doesn't go to KSTEP. Yes it is preferred to use an isolated supply but not absolutely necessary if you don't mind having a common ground.

Regards

[John Coloccia]

Question about the grounding architecture. I believe that all of the "GND" connections, including the eyelet locations on the KStep/KFlop board, are all tied together and are actually the +5 PS return (except for the motor PS ground which is isolated and is the motor PS return). I think it's NOT intended for the boards to be tied to chassis ground through the GND eyelet but rather should be isolated from the chassis. Then it's sufficient to power the KStep/KFlop entirely through the molex on KFlop and that +5/Return completely satisfies the logic requirements, and the motor +48/Return completely satisfies the driver requirements. Does all of that sound correct?

[TomKerekes]

Hi John,

Yes I agree. If you run in non-isolated mode all you need to do is:

apply +5V/GND (or run USB power if IO requirements are low) to KFLOP

apply VBB Motor Power and GND

This diagram may help.

This KSTEP Video shows only USB Power and Motor Power connections.

As you state I don't think it is a good idea to tie DC ground to chassis ground multiple places as this will potentially create ground loops. The mounting holes on KFLOP and KSTEP are isolated. Only the holes marked GND have a DC ground connection.

HTH
Regards

[Al_The_Man]

There is a alternative practice of equi-potential bonding, where all ground points are take to a chassis or a star point ground.
http://www.automation.siemens.com/do.../emv_r.pdf?p=1
In the video using a lap-top the P.C. P.S. is most likely isolated, but on a tower or desk top, the USB 5v common will be at Earth ground potential via the M.B. ground.
Al.

[John Coloccia]

Alright, one last question and I think I'm good to go here. I'm hooking it up to a Hitachi Inverter (WJ200) that I'd like to switch on and off. The input can be configured as sourcing and the inverter supplies 24V logic power. I was thinking I could use the relay driver on KStep directly to drive the inputs on the inverter (i.e. 24V from inverter to Relay0+, Relay0- to inverter input). I wonder if you think this will work OK? I also wonder if I did this if it would be prudent to maybe put a 10K pull down on Relay0-?

[TomKerekes]

Hi John,

That should work. I don't think a pull down will be necessary.

Regards

[John Coloccia]

Thanks again for the help, Tom.

Al: my concern grounding everything together was the potential for introducing a lot of noise, especially since I have two different supplies, one of which isn't even mounted in the box. I thought in this case the best approach would be to ground the chassis, and then just let the boards sit at whatever reference they get from the PS, whatever that may be, instead of trying to reference everything to ground. I'm using a tiny little micro ATX PS. I didn't bother to check if the returns are tied to chassis ground. I'll check that out today just to satisfy my curiosity.

[Al_The_Man]

There are two schools of thought, total isolation of all supplies or make all common to the Earth Ground star point, I have always followed the latter on P.C. bases systems.
It is usually not the P.S. that is grounded, but the M.B. ground plane.
Older P.C.s often had nylon stand offs for the M.B. but modern design is usually 6 or 7 ground plane screws that mount to the earth grounded frame of the P.C cabinet.
Al.

[John Coloccia]

Just a quick FYI. My particular PC power supply case is tied to the returns, and thus everything is tied to the chassis. This is good because the frame will eventually be bonded too and a touch plate will work. The only thing that's isolated it the stepper PS, so life is good Everything logic oriented is...well...logical!

[John Coloccia]

Well, everything is going very nicely. I have motors moving and generally everything seems happy except for one thing:

When I move an axis, it will move forward, and then it will move backwards a bit. If I decelerate slowly with the gamepad, it doesn't do it, but if I decelerate quickly (or use GUI to jog) it will. It's almost like there's a filter running and some negative term is winding up, but I don't think anything is running.

Anyhow, I'm sure it's something stupid. If anyone has any ideas, do tell

Thanks again for all the help.

edit: Figured it out. Have to define the axis, even just for jogging, or you get strange behavior.

[John Coloccia]

Question about currents and limiting. I was tuning the axis a bit and decided to check the current. I have them all set with one jumper in "H" for 3.1A limiting. When I check the current on the 48V into the KStep, it only draws about 300ma per motor. It also does a weird thing where if it skips steps with one axis plugged in, the power supply makes a clicking noise as though it had tripped the crow bar, but with two motors plugged in it doesn't. Some sort of flyback issue? I don't think that's really an problem, but I just thought I'd mention it to get your wheels turning.

For the record, I was testing the +V side of 48V into the KStep with a Fluke 85. I also tested the PS itself to make sure it wasn't sagging, and it's rock steady at 48V. I guess I should test the 48V on the supply when it skips steps to confirm that it really is or isn't crowbaring.

Thanks for any insight, Tom

edit: FWIW, one motor plugged in is definitely causing the supply to crowbar. Min/Max shows an excursion to 50V and min of 0V. Times like this make me miss have a scope around, but anyhow there it is.

Again, I don't know that any of it is really a problem. I'd just like to understand what's going on

[TomKerekes]

Hi John,

The low current from the Supply is normal. KSTEP is a switching amplifier so it is basically sometimes pulling 3.1A (maybe 53% of the time) from the power supply and sometime Pushing 3.1A (maybe 47% of the time) back into the power supply. The Capacitance and your meter averages things out to see an average drain of 0.3A.

Another way to look at it is from a power/conservation of energy perspective. I don’t know the resistance of your Motors but if they are for example 1.5 ohm. Then the heat dissipated at 3.1A would be about 14 Watts. 14W from a 48V Supply requires 0.3A.

KSTEP is very efficient as it uses 0.017 ohm MOSFETS so virtually all the of the Power Supply energy goes into the motor as heat or useful work. When the motor starts spinning at high speed under load the Power Supply will see more power/current drain.

Similarly when a spinning motor stops (including a stall) the kinetic energy of the spinning motor needs to go somewhere. Normally this energy is pushed back into the Power Supply. This is often referred to as regenerative braking. This can cause the Power Supply voltage to spike upward and either trip a fault in the supply or cause damage to the supply or KSTEP. KSTEP has a built in voltage clamping capability that is capable of momentarily dissipating huge amounts of power to avoid this problem. The voltage clamping is fixed at 49V (but can be lowered by adding jumpers). So I don’t know why your supply appears to be faulting unless it doesn’t like a spike of 49V. With another motor connected which is drawing power the amount of energy pushed back into the supply will be reduced. This is probably why you don’t observe the problem with multiple motors connected.

Please make sure you do not have any voltage clamping jumpers installed. If you do the clamp circuit will try to clamp the supply below 48V, draw a huge current, trip the resettable fuse, and disable the clamping feature.

HTH
Regards

[John Coloccia]

Hi John,

The low current from the Supply is normal. KSTEP is a switching amplifier so it is basically sometimes pulling 3.1A (maybe 53% of the time) from the power supply and sometime Pushing 3.1A (maybe 47% of the time) back into the power supply. The Capacitance and your meter averages things out to see an average drain of 0.3A.

Another way to look at it is from a power/conservation of energy perspective. I don’t know the resistance of your Motors but if they are for example 1.5 ohm. Then the heat dissipated at 3.1A would be about 14 Watts. 14W from a 48V Supply requires 0.3A.

KSTEP is very efficient as it uses 0.017 ohm MOSFETS so virtually all the of the Power Supply energy goes into the motor as heat or useful work. When the motor starts spinning at high speed under load the Power Supply will see more power/current drain.

Similarly when a spinning motor stops (including a stall) the kinetic energy of the spinning motor needs to go somewhere. Normally this energy is pushed back into the Power Supply. This is often referred to as regenerative braking. This can cause the Power Supply voltage to spike upward and either trip a fault in the supply or cause damage to the supply or KSTEP. KSTEP has a built in voltage clamping capability that is capable of momentarily dissipating huge amounts of power to avoid this problem. The voltage clamping is fixed at 49V (but can be lowered by adding jumpers). So I don’t know why your supply appears to be faulting unless it doesn’t like a spike of 49V. With another motor connected which is drawing power the amount of energy pushed back into the supply will be reduced. This is probably why you don’t observe the problem with multiple motors connected.

Please make sure you do not have any voltage clamping jumpers installed. If you do the clamp circuit will try to clamp the supply below 48V, draw a huge current, trip the resettable fuse, and disable the clamping feature.

HTH
Regards

Thanks Tom. That explains it completely. Like I said, I do miss having a scope around. 2 seconds with that would have shown me exactly what was happening.

Yep, all my clamping jumpers are out. Things appeared to generally be operating properly so I knew there was some explanation for what I was seeing. Honestly, I'm not sure why my supply is tripping either. I generally expect to see them trip when the current limit is reached, not when the voltage is a little too high. I think my 85 has a 100ms sample on Min/Max, so maybe it's seeing some huge transient that's setting it off. Anyhow, it never happens with more than one motor plugged in so I'll just go back to ignoring it.

Thanks again. It's much appreciated. This whole integration is going very smoothly. Nice.

[sahilkit]

Hello everyone,

I have finished testing my K-Flop, K-Step, power supply, motors and want to build everything into a nice cabinet, need help/advice on the grounding of P.S, boards etc.

here are some more questions.

1. What is the maximum amperage drawn by the Isolated 5volts J6 or JR1 connectors ? will a 5volt 2amp P.S do ?
2. What is the maximum amperage drawn by K-Flop alone at 5volts ? will a 5volt 1amp p.s do ? when using only JP36 connector ?
3. Will a common 24 volts supply work for relay drive and input connections on JP33 connector ? and 24volt 3amp P.S do ?

Bye
Sahil

[TomKerekes]

Hi Sohil,

I think the requirements for each of those supplies would be less than 1Amp.

Regards

[sahilkit]

It's SAHIL,

Thanks Tom, the reason i asked for amperage was 2.5Amp peak current consumption was mentioned in documentation of K-Flop.

[TomKerekes]

Hi Sahil (sorry),

That 2.5A is quite conservative and assumes every possible option board and IO loading. Also 2.5A can be obtained very inexpensively with something like:

RS-15-5, Mean Well RS-15-5

Regards