RapidChange are working on Mach4 at the moment (early alpha I think), but they have a stack of controllers to support so I think it’ll be a bit longer. I’m just trying individual tool change macros at the moment.
The spindle is fine at 1000rpm for short periods (it’s a cooling issue at low rpm, so as long as it doesn’t run long enough for it to get hot, it’s fine. A tool change takes seconds)
I have my RapidChange ATC installed and working with the AVID as far as the IR sensor and Lid control goes. RC is working on the macro’s right now for the tool change since they haven’t supported Mach4 up to now.
As far as the spindle speed, you can set that lower. I have run mine at 500rpm for drilling operations. Depending on the age of your system, you may only have to change the min speed setting in Mach4, or you may also have to set that in the VFD parameters.
AVID sets them higher because if you try to do heavy cutting at those low speeds you could overheat your spindle since its air cooled and the fan isn’t moving fast. However, for short operations like drilling or spinning on/off the collet nut its not really a concern.
I bet you could adapt my screen set to work with it. It just needs no “slide” distance, but is basically the same as the other styles…although, it needs some specific time/duration to lock the collet, right?
Ya, they have worked up a standard approach to spinning up the spindle and dropping it into the collet to attach it and remove it.
I was going to make a screen, but they want to make one thats their own product so they can support it.
I am going to help them get theirs working so I can test the whole thing first. I’m really curious to see how reliable it is to just spin on/off the collet like that. The ATC has a ball bearing lock mechnanism to grab the collet.
In the end I’m going to look at your screen, and theirs, and then probably combine features of both plus my own stuff (just because its fun :-)).
Yeah! I’d probably do the same thing. The code I wrote is pretty modular, so you should be able to easily plug it into somewhere else. One thing I started doing was to not write any real code in the screen script, and put it all into modules. This made debugging and portability easier.
Thanks.
I already change the min spindle rpm in the machine.ini to 500 rpm (2.2Kw spindle + avid VFD)
Should I also change the VFD parameters ?
If so, do you know where i can find the procedure ?
So gratefull for your help !
AVID has the procedure for changing parameters on their website for the two spindle types, and you can download the VFD manual from the web to find the actual parameters for spindle min speed.
Thanks for all the good information. While i currently have it working as described one bit of caution on the use of the spec’ed relay module, the contacts on the relay used are rated at 10A and as such are made of a hard material that is meant to be kept clean of oxide by arcing. The VFD spec shows the inputs are whetted with a resistor that allows about 16mA of current to flow, this is not enough to keep the contacts clean of oxides and because the contacts are hard to stand up to the arcing the oxide does not shatter on contact closure. The relay will eventually fail to close because of this. Relays for low current (2A or less) use gold plated contacts, the gold is malleable so the act of closing the contact shatters the oxide layer so the relay continues to work properly even with low current. Unfortunately i have been unable to find a readily available relay module with the appropriate relay so will be taking a different track and using an optoisolator module. I will post details once i have thst in place hopefully this coming week.
You are correct, this is not a small signal relay, and a similar module with a better relay is hard to find. I could have probably just wired in a better relay by soldering, but I really wanted one with screw terminals. I guess it would be easy to use the driver on the module and just replace the relay itself with a more suitable relay too.
Opto isolators can be problematic in this application as well due to their body diodes. You will need one that doesn’t allow reverse current. If you find one that works, please share it with us, the first one I tried didn’t work.
or, easy enough to swap the relay on the module out
Ok, the relay has been bothering me for a while, because as bockert said, this isn’t really the right relay for a small signal. I went looking on Amazon again for a nice solid state relay module that has the back to back FETs in it for proper isolation and I found a good one, and verified it works today.
This is a nice one for CNC work because it also has indicator lights on it. The input, power and outputs are fully isolated from each other. Outputs can handle up to 60V, and the power supply can run from 5V to 32v. A nice solid state signal relay for CNC work. It comes in 1, 2, 3, and 4 channel versions as well.
This is the module https://www.amazon.com/dp/B072Z628LP?psc=1&ref=ppx_yo2ov_dt_b_product_details
I updated my document accordingly: Public CNC Docs - Google Drive
Thanks for the impetus to fix this Bill 
Hi Jim, nice solution. I went a slightly different route because I had the VFD information in hand. The inputs on the spindle are whetted to 24V through a resistor that limits the current to 16mA. Because of that most optoisolators can handle the connection directly without the need for the transistors between.
I ordered the following
https://www.amazon.com/dp/B01L1OI1HC.
These are actually set up for 12V input however by changing R1 to 260 Ohms they work just fine at 5V. That resistance sets the current through the opto input to around 10ma which is well withing the drive capability of outputs on the ESS. The EL817 opto output is rated at 50ma and 36V so well within the needs to drive the VFD.
The connections are
Input - to GND on the ESS
Input+ to Output on the ESS
Output GND to VFD DCM
Output OUT to VFD MI2
Note Output VCC is not connected as the VFD input is self whetted so this solution has the advantage of not needing any supply voltage.
That similar to the one I tried first. I had problems with it conducting backwards through the body diode, but in hindsight, I should have just reversed the connections on the output side and it probably would have worked.
Jim
I do not believe the EL817 has a diode on the output, it is strictly a photo NPN junction and as such will only work if connected with the load on the collector and ground on the emitter. If hooked up backwards it would not work at all.