Reply with QuoteA question I have is do the drives have a drive-disable input? generally this turns off the output FET's or IGBT's, as the case may be.
Al.
E-STOP Circuit for the UHU
Hello all,
I started to wire my e-stop circuit which will handle my 4 axis setup and just wanted to share my ideas and see your opinion on whether this will work or not.
So I have a 24V relay that is energized through all the E-stop buttons, limit switches and the charge pump from the computer. When I have a normal condition the relay is switched ON. A NC contact on the relay is wired from the UHU E-STOP IN pin through a diode to ground. All 4 UHU's are wired in a loop, so one's OUT pin goes to other's IN pin. This halts all the boards when a single one errors.
Another pin from the relay is used to energize a magnetic contactor, which puts resistive load directly on the motor leads. When the contactor is switched ON the resistors are disconnected from the outputs.
Another contactor which switches the primary of the power supply is also wired to the E-stop relay. When it is OFF it cuts the primary and puts a load on the capacitor bank. So far so good.
My question is, I know that when the UHU chip is on red /the error led is ON/ the output pulses are cut, and the motor is let to spin free till it stops. Can I safely put this load resistors on the output without damaging the drive, and eventually stop the motor fast enough? My motors accept 100A peak current, so with a power supply of 65V and full speed if I put 2 Ohm resistors on the leads it gives something like 32.5A in the first moment and descending after that.
In my understanding, there should be no problem to do this but I wanted to be sure before I blow some FET's.
Sorry for the text explanation, I will try to draw a schematic of this later to be more specific.
Thanks, Todor
EDIT: I am using the OLD UHU if this matters.
Similar Threads:
A question I have is do the drives have a drive-disable input? generally this turns off the output FET's or IGBT's, as the case may be.
Al.
CNC, Mechatronics Integration and Custom Machine Design
“Logic will get you from A to B. Imagination will take you everywhere.”
Albert E.
The error condition is triggered when a low state is applied to the E-STOP IN pin of the controller or when the maximum deviation from the position is exceeded. This switches off the output and the motor free-spins after that. No braking action is made as I recall.
Thanks, Todor
Todor,
If you can be SURE that the output stage disables BEFORE you connect the resistor across the output then it will probably work. Otherwise there's a risk of damaging the MOSFETs since the current limit circuit doesn't operate properly on a resistive load. (There's no inductance to limit the rise time of the current so the limit circuit don't have any time to react, if the MOSFETs can stand the 32.5A then I think it should be OK)
You say that you switch off the primary of the powersupply and connects a resisotr across it to blead it out. Why not make the resistance of this resistor low enough to break the motors? The body diodes of the MOSFETs will conduct current from the motor back to the resistor so it should work just as well. As long as you don't have too low resistance (that'll damage the diodes) it should work fine.
/Henrik.
Hi,
I am working with the IRFP150N, which has a body diode rating of 42A. Additional to that there are external BYV27/200's , which are some 6 times faster and handle the voltage spikes better than the internals when working.
So I guess I will try your approach first and if I am satisfied with the results will work with it. So with a capacitor bank of 4x6800mfd + 8x1000mfd on the boards = 35200mfd at 65V. 2 Ohms bleeder will almost fully discharge it within 0.25sec. As long as the online calculator I found is true.
The problem is that my rapids are 10m/sec, which gives 166mm/sec. So for a time of 0.5sec it will move 83mm, but my hard stop is some 25-30mm away from the limit switches.I guess I have to use smaller resistor.
Todor
EDIT: At that particular rapid speed I have to stop the motors for a maximum of 0.12sec as they don't hit the metal. 2 revolutions here gives me 20mm of travel.
Hi Todor,
You should have a look at the maximum discharge current of the capacitors as well, they may not like being discharged TOO aggressively. Also, remember that you would get 83mm travel in 0.5 seconds IF the voltage remained at 65V - which it doesn't.
You have a delicate problem there, breaking "gently" risks damaging the machine, breaking hard risks damaging the drives.... Perhaps you need to consider moving the limit switches "inwards" a bit to get a larger safety zone and/or add some shock absorbers at the definitive end.
/Henrik.
Hi,
I was out of town for a couple of days so could not respond. Yes, I've got 65V only at the very beginning, It goes down by exponential curve after that. But you know things sometimes differ in theory and in practice.
As a start I will use higher value resistors on the output and on the capacitor bank, so I don't get any high enough currents that could destroy something on the drives. Then I will try it out and see how it performs and lower the ohms after that if necessary.
Apart from that I never realized how much different materials, instruments, apparatus, bolts and nuts are involved in the making of a cnc control box. I ended up with all this stuff laying all around just not to have enough space to turn.
Here is the outside view of my electronics box, the computer is still away, and the back and side panels are still not mounted. All is done with plywood and covered with PVC foil. Sorry about the quality.
Todor
Todor,
That looks nice! May we see a picture of the machine as well? Please?
/Henrik.
The machine is almost done, we are doing the spindle setup right now, motor from a 2KW angle grinder and custom made tool holder below it. It will mill mainly wood, ply, MDF and other soft stuff.
I will photo it when we mount the spindle and most of all when we clean the mess around there. It is not to be seen like this now.
It took more time than I thought it will, because I designed the BOB, the relays board and all the wiring to the computer myself, PCB manufacturing and all. Not to mention I intended to use proximity switches for homing and index "C" pulse from the encoders to do it. Now I think that for a woodworking mill a simple limit switch like my old machine would do the same, but I wanted to do it like the BIG guys.
Todor
So I draw this schematic, which covers the power stage on/off switching and the E-STOP.
TIME1&2 are variable time relays set to 1sec. 110V uninsolated transformer is used only for them and the power contactors. +18V UHU is going to the 7812 regulators. I am still not sure about the values of shunt resistors, that's why there are none.
Advise on errors and improvements are welcome.
Todor
Hello All,
So I see very little activity here, anyway I will share my progress on the control panel. I have almost finished everything, but had to make some changes to the schematic.
I wired a NO contact of the E-STOP relay in the LIMITs chain so that when an E-stop is triggered it stays in this state forever. I have to push the override button to switch it on. I added another relay that tells the E-STOP circuit that I have mains. If I have blackouts or deep brownouts I got an E-STOP condition.
When I firstly reviewed Kreutz's HP schematic I wondered why he put the optoisolation on the E-STOP input/output. Today I no longer do that. I have formerly wired a loop that covers the 4 drives, but had frequent RESETs of the UHU chip when running. Very annoying and took me some time to realize what was the reason. As soon I disconnected the E-STOP IN pins, all problems went away. I will have to figure out something to make it work as I wanted in the first place.
I tested the limit switches. It takes about 15-22mm so that X axis stops from a speed of 5m/min with a 5 Ohm resistor across the capacitor bank . Not very satisfactory but good as a start.
Homing with the index pulse and the proximity switches works perfectly. Tomorrow I will put some dial indicator to see what is the repeatability. The spindle is still not ready but I am eager to see it milling.
Todor
