Originally Posted by
Mariss Freimanis
G540 next revision (REV 2). I took some of the advice to heart and added a few additional goodies:
1) DISABLE will be renamed ENABLE. The ENABLE input must be shorted to GND to enable the G540. If ENABLE is left open then the G540 shuts down.
2) A watchdog timer (charge pump) is added. No 25kHz on the DB25 pin 16, no run.
3) The +5VDC input from the PC is eliminated. An isolated internal DC to DC converter coughs up the +5VDC at 10mA need for parallel port input pullup resistors.
4) Two general purpose outputs are eliminated, serviced by DB25 pins 14 and 16 in the current G540 REV1. Pin 16 is now the dedicated "charge pump" output and the use of pin 16 is discussed next (5).
5) An optoisolated VFD interface is added. The interface uses 3 terminal positions; VFD +V, VFD -V and VFD OUT. This circuit is completely isolated from the rest of the G540 to a minimum 300V breakdown voltage.
My understanding is most VFDs provide a remote potentiometer connection of 10VDC at 10mA. VFD +V goes to the '+' side of the pot connection, VFD -V goes to the '-' side and VFD OUT goes to the pot wiper or input connection. The output voltage range is from 0V to whatever '+' is (this must not be more than +12VDC). The VDF signal must be a PWM output on the DB25 pin 14. The -3db bandwidth is 5Hz.
6) A FAULT output back to the PC has been added. This output reads as a '1' when the G540 has faulted out (shut down) due to a short-circuit on a motor lead or when the ENABLE input to GND is disconnected (E-Stop). It is on the DB25 pin 15.
7) Internal ultra fast blow fuse protection. This is to help with the power supply polarity reversal thingy.:-)
8) We will be including 4 mating male DB9 connectors (solder cup type) and 4 sets of DB9 connector backshells with each G540 to ease the user's motor cable fabrication work.
Comments please.
Mariss
1) Sounds good. One of the uses for this pin is to manually disable the drives, so that the axes positions can be tweeked by hand. This is not always desirable to have Mach see this as an EStop.
2). Not sure why you chose 25KHz? I think that the standard chargepump frequency is 12.5HKz. Maybe it is now configurable. Also, you should consider providing a jumper so that the system can operate with a logic signal rather than a frequency (Not all software can produce a charge pump).
3). :-) I have always failed to understand why BOB manufactures don't do this, relying on the user to supply another powersupply source.
4). OK
5). VFDs usually have a +10V 10-20mA supply for this purpose. The more common and cheaper speed controllers that will most likely be used with the G540 do not. They use a Zener diode shunt regulator to provide power for the manual Potentiometer.
The kicker is that they can only supply a few mA, certainly not 10mA. Basically, just enough for the 5K pot (3mA), even less possibly for those designed for a 10K pot. If you draw more than the 3mA, the supply to your converter circuit will sag (the amount of sag varies with the active impedence of your circuit. The result is a non-linear speed response and a reduced maximum speed -Not good)
Also, most of them provide 15Vdc such as the KBIC-120 controllers as used in the sherline products, so your 12V limit will not be sufficient. Many of these controllers can be controlled via a 0-10V signal referenced to the -ve pot terminal. Some still need the pot (or a resistor) attached (without the wiper connected).
Basically, the bottom line is that for reliable operation, you need another dc/dc converter (12V-15V), suppling the isolated circuitry rather than relying on the speed controller to provide it.
The Digispeeds DC-06, now have an optional plugin DC/Dc converter for this purpose, If it is to be used with a VFD the converter is not needed, but for cheaper controllers it is recommended.
6). I don't know if you have addressed it, but make sure that you avoid a race condition between the fault input and the chargepump. If the chargepump is not present the fault is activated, which disables the charge pump, which activates the fault line........
7). Good
8). Good.
Well you did ask for comments.... :-)
Cheers,
Peter.
-------------------------------------------------
Homann Designs - http://www.homanndesigns.com/store