[GeckoSub]

Buy the ready made cables!

Usually:
Step pulse, ideally differential, therefore 2 wires
Direction pulse, ideally differential, therefore 2 wires
24v or 0v (depending on preference/wiring) common for inputs and outputs
Estop signal into an input - estop should ideally physically cut motor power input, but drive can actually actively stop motor (when told to do so) faster than cutting the drive motor power input will (large capacitors in the drive)
You can also feed in limit switches

Thanks Pippin,
Good to hear a yes vote so will go ahead with these
They are in the shopping cart but haven't decided on a length, yet though I think 0.5m should suffice (next up is 1m).

[joeavaerage]

Hi,
I bought my Delta servo new and so they came with the plug ready to solder up.

I just used two light weight signal cables, not shielded or anything and used as many cores (8) as I needed.
Takes about an hour to solder in the required wires and assemble the plug....not too bad.

Step= 2 (differential)
Dir=2 (differential)
Alarm=1
Reset=1
Enable=1
COM =1

Total=8.

I used the CNC controller to monitor the limit switches, rather than hook them direct to the drive. No particular advantage either way. I also relied on the Enable line going
inactive to stop the servo rather than having an emergency stop signal. Note Delta servos are equipped to handle a dedicated emergency stop signal if you want it.

I was trying to limit the number of wires so that I could terminate the other end in a 3.81mm pitch 8 way plug, which in turn plugs into the corresponding 8way socket on my breakout board.
One socket per servo...easy.

Craig

[GeckoSub]

Hi,
I bought my Delta servo new and so they came with the plug ready to solder up.

I just used two light weight signal cables, not shielded or anything and used as many cores (8) as I needed.
Takes about an hour to solder in the required wires and assemble the plug....not too bad.

Step= 2 (differential)
Dir=2 (differential)
Alarm=1
Reset=1
Enable=1
COM =1

Total=8.

I used the CNC controller to monitor the limit switches, rather than hook them direct to the drive. No particular advantage either way. I also relied on the Enable line going
inactive to stop the servo rather than having an emergency stop signal. Note Delta servos are equipped to handle a dedicated emergency stop signal if you want it.

I was trying to limit the number of wires so that I could terminate the other end in a 3.81mm pitch 8 way plug, which in turn plugs into the corresponding 8way socket on my breakout board.
One socket per servo...easy.

Craig

Hi Craig,
That BoB looks sweet, well done

Gotcha on the limit switches, my gut feeling is I will let the controller handle it, too.

Which reminds me of an aside that I wanted to mention a while back when we talked about the differences between B2 and the newer B3 and A2/A3 series. I have in my notes somewhere that B3 can actually home to hardstops - all handled by the drive. Just wanted to mention it in case someone someday stumble across this thread and was wondering what some of the differences were. I think we already talked about how B2 can't handle encoder inputs through the drives but some of the others can. (Keep in mind, my notes are not the best, so anyone reading this down the line, please do double check this).

[GeckoSub]

Early Test of Spindle
I hooked up the Chinese spindle (24K RPM, 4 poles, 800Hz, 2.2kw) to the high speed Delta MS300 VFD and after going over the parameters a bunch of times, I dared pressing "run".
I don't have water cooling set up yet so only ran the spindle 15-20 secs at a time but I already have a few questions:

- It runs in reverse despite following the wiring diagram from the (Chinese manufaturer), haha. Is it true I can just swap any of the three non-ground wires to make it run in the opposite direction?
- Do I ground the shield in the cable? If so, ground it at both ends? The latter question may sound silly but it's a left over that stuck in my head from an old thread where someone said to only ground it at one end...
- I have ramp speed set to 5 secs for now, does that sound about right? I don't have a breaking resistor.
- The spindle is a 10A beefy one, but it draws only 2.2A as it ramps up and down and about 1.6A at full speed - that's what it says on the VFD readout. Does that sound about right for a no-load scenario?
- There's a min. frequency setting on the VFD but it's set to 0 as default. Supposedly these spindles don't like to run slow so does it make sense to change this setting to something lower so I don't accidentally end up running it too slow for too long someday?

[joeavaerage]

Hi,

Yes you can reverse the direction by swapping any two of the three phase wires. Most VFDs allow you to swap direction programmatically as well. That would allow you to
ensure that FWD on the control panel is rotating correctly for right-handed tools.

5 seconds is a generous ramp up time. It means that only modest currents will be drawn when accelerating up to speed, but more importantly diminish the voltage surge that occurs
when decelerating. I have mine set at 6 seconds....been like that for over ten years. Its enough. If you are doing production and have an ATC the 1 second or less would be common,
but for that you'll want a braking resistor!

1.6A at full speed no load sounds about right. The spindle is an induction motor. The 'induction' part comes from the fact the a proportion of the stator excitation (from the VFD) 'induces' current in
the rotor, and is how an otherwise un-magnetised rotor becomes magnetised. Thus you will need some current, maybe 0.5A, maybe 0.8A to magnetise the rotor and the balance to overcome windage
and friction.

Even water cooled spindles do not like being run at part speeds. If you were being conservative you might limit the slow speed to half rated speed (12000rpm), that would go a long way to protect your
spindle. You could reduce that to 1/3 or 1/4 speed (8000rpm-6000rpm), but be aware that if you take heavy cuts at these low spindle speeds you risk damage to your spindle.
I have an 800W air cooled spindle and so I'm doubly careful not to run below 12000rpm....but sometimes I have to. I have a carbide slitting saw, and very useful it is. Its only 0.25mm thick
and 27mm diameter. Being as thin as it is make for a very VERY narrow kerf, but also make it very fragile. I doubt it would survive the centrifugal forces of spinning at 24000rpm, so I run it at 6000rpm.
I don't use it a lot, and very carefully and quietly when I do, and for the brief low power use at 6000rpm has caused no harm to my spindle and surprise surprise the slitting saw is still intact five years
after I bought it!!

Craig

[GeckoSub]

Thanks so much Craig
Will swap the wires around and keep the ramp time as is for now.

Also, good to hear your experience with the slitting saw. I am still only dabbling in CAM in my head but for one of the parts I need to make, there seems to be a case for a thin slitting saw, too.

[edit] I found the setting in the VFD that changes the direction instead of changing wires. Delta calls it "Output Phase Order Selection":

[hanermo]

I suspect that the low-power slitting saw will be a non-issue re:spindle lifetime.

Commercial VMCs, large, typically only use about 2-3-4 amps at 3-phase 440, except in getting up to spindle speeds.
They use the full 40 HP/25kW to get upto the max rpm as fast as possible, about 3 secs, and after that only a few kW in a 25 kW machine.

But they have very rigid class 1 screws, tensioned sometimes, and tensioned preloaded rails and carriages.
This reduces the hammering on the spindle that is likely to reduce the lifetime.
A commercial haas spindle can typically expect to last 10.000 work hours cutting steel.

[GeckoSub]

Brake Servo For the Z-Axis - What Do I Need To Make This Work?
I have Delta B2 servos but I am honestly not strong in the electrical department. The manual says 24V will disable the brake and there's also a schematic showing a solenoid relay and diode.
So, I guess I need to buy that? The manual also says the brake only draws ~6w.
I guess I could find a relay with that info but no idea of how to spec diodes.



How about programming? Does the controller disable/enable the brake or does that happen automatically inside the servo drive?

[peteeng]

Hi Gecko - I use brakes on Z axis. I use a 24V power supply for various things so take a lead to the brake. If the main power is on the 24V power is on therefore the brake is off. If the power goes down the brake is on. Peter

[GeckoSub]

Hi Peter,
Good to hear from you
Just to be sure: You don't run the 24V through a relay controlled by the CNC controller, you just run it "dumb" - as in only on when the whole circuit is on?

I kept thinking that the brake would engage and disengage a lot in between moves but of course, that's not needed - the servo when enabled/powered on, even when not moving, will hold the Z just fine. So, the brake only needs to be engaged (24V off) when the servo is powered down.

So, actually, I don't think I need the relay for this. Just keep it simple. So, you don't use the BRK signal in the servo drive for anything to do with the controller either, right?

[peteeng]

Hi Gecko - As you say its a dumb connection. But works. I have had trouble with one PSU that had a bit of latency so the Z dropped a little. Its voltage didn't drop far enough (when powered down) for about 0.5 sec. So I swapped the PS and it was then instant... Peter

[GeckoSub]

Hi Gecko - As you say its a dumb connection. But works. I have had trouble with one PSU that had a bit of latency so the Z dropped a little. Its voltage didn't drop far enough (when powered down) for about 0.5 sec. So I swapped the PS and it was then instant... Peter

Perfect - I prefer simple - and will keep that issue in mind if it doesn't work for me the first time around.
The only other issue I can foresee is if my servo drives take a lot longer to boot up than yours in which case the Z will have time to drop.

[peteeng]

Hi Gecko - On one machine it dropped and I used relays, different power supplies, timers etc. Turned out to be a sticky brake!! Argggg Peter

[joeavaerage]

Hi,
I think you are making a mountain out of a molehill.

The brake on a Delta B2 servo is just that.....an electromagnetic brake on the end of the servo motor. It has two wires....that's it. The servo drive does not control the brake,
unless you make it do so. The 24VDC flyback power supply inside the drive would be overtaxed to hold the brake off....do not use it!. You should have a separate supply.
Its would be nice if it were regulated but it does not need to be regulated. The brake will release if 18VDC is applied, but for better surety of release and minimal residual drag 20VDC is better.
My brake draws about 300mA at 30VDC and about 225mA at 24VDC. Any DC voltage between 20VDC and 30VDC will be fine.

I use the enable signal to the servo to operate the brake. All servos will require an Enable input from your control system. If the servo is enabled the shaft will hold itself in position
unless commanded to move, and therefore the brake can be energised, ie released, without any issue. When the servo is disabled the brake de-energises and then locks on preventing shaft rotation,
great new for a Z axis servo as the weight of the spindle will not cause the axis to sag into the work zone....where you are loading another part into the vice.

There will be a few milliseconds between when a servo is disabled and when the brake locks on. I personally have had no issues with that delay. There is however in the servo drive a couple of parameters
that can be programmed that accommodate that delay. The you can program one of the drive outputs to operate the brake via a relay or an amplifier as you have posted. To be honest I don't think its required,
but its your choice.

My breakout board has an output to the Z axis brake, and it is driven (logically speaking) by the Enable signal. I use a 1A Darlington transistor (and flyback diode) to switch the 24VDC (nom) supply.

Craig

[joeavaerage]

Hi,
by the way; in the circuit diagram you have posted that uses a relay I note that you don't have a flyback diode, that is very poor.
The inductance of the brake will cause a very distinct arc at the relay contacts at the time the contacts open. They will not last long in that service.
You need a flyback diode, also called a catch diode, and if you were being really picky a low value ceramic capacitor as well.

Craig

[GeckoSub]

Hi,
by the way; in the circuit diagram you have posted that uses a relay I note that you don't have a flyback diode, that is very poor.
The inductance of the brake will cause a very distinct arc at the relay contacts at the time the contacts open. They will not last long in that service.
You need a flyback diode, also called a catch diode, and if you were being really picky a low value ceramic capacitor as well.

Craig

On the way out the door but will digest your posts when I get back.
But the wiring diagram is from the Delta manual. It does have a diode on it - but not sure if it's a "flyback diode" or a regular diode since I am clueless when it comes to these things

That said, when I was looking at relays, it seemed some of them had a diode built in?

[GeckoSub]

Z-Brake (and Relays) Cont.
I do have a designated PSU for the brake on the Z-servo, it literally just arrived with a friend from China (the end mills are for a friend):


Plenty powerful at ~35w as the brake only draws ~6w according to the manual.

First, I will try to run it dumb like Peter does, but I also think tying it to the enable signal sounds like a smart thing to do.
So, I think that I should still order a few relays since I can use them for other things, too like a mister.
I came across what's dubbed as PLC Relay Modules and they are very affordable and come in many different sizes in terms of how many relays are in the module. I thought they make sense as a nicely packaged way to get into this? Judging from the schematics, there's some sort of diode setup in there:


Same with this single relay, it seems to have a diode, too:


Will these work as is so I don't have to mess about with soldering diodes myself?

[GeckoSub]

Brake Z-Servo Tested
Just did a quick and simple jog test on the last Delta B2 servo of the four I bought used in China. I couldn't test this one until now since I was missing a 24V power supply to disengage the brake.



Well, it works. It does have a different sound than the non-brake servos when it starts and stops, like a mechanical clicking sound. Even with 24V always fed to the brake. Maybe my mind is playing a trick on me but I feel like I have read in a manual or online somewhere that this is normal but can't find it now.

As for hooking it up "dumb" with the 24V PSU just turning on when the servo drive turns on, there's a bit of a delay. As much as 2 secs where the brake is disengaged before the servo is powered up and holding station. On turning off, it's much shorter, something like 0.5 sec. When I try to turn the axle as fast as I can with my fingers, I can get about 1-1.5 turns during the turn on phase and about half a revolution when turning off.
I obviously have no idea how fast a Z assembly will fall but even if it's the same speed as my hand turning, the Z will drop by less than 10mm.
So, yeah, dumb is good enough for now and then if I want a small project later on or the drop is jarring I will figure out how to have either the drive control it or the CNC controller.

Still, if you have feedback on the relays in the above post, let me know as they would be good to have and everything takes months to get here.

Overall, I seem to have had very good luck on buying secondhand in terms of the servos and VFDs. I am 99% sure the high speed VFD hasn't actually ever been used. No dust, no scratches and the screws for the terminals were done up super hard and even between all of them and there was a warning sticker covering the terminals that would have been hard to keep in place if one had actually had to connect cables.
The other, normal speed VFD certainly saw some use, it's chuck full of dust inside, so I want to give that a good clean before I test it. But all the four servos and cables seem fine so far.

[joeavaerage]

Hi,

As for hooking it up "dumb" with the 24V PSU just turning on when the servo drive turns on, there's a bit of a delay. As much as 2 secs where the brake is disengaged before the servo is powered up and holding station. On turning off, it's much shorter, something like 0.5 sec. When I try to turn the axle as fast as I can with my fingers, I can get about 1-1.5 turns during the turn on phase and about half a revolution when turning off.

Cant imagine what's going on. My brake locks on within a few milliseconds, I've never had the servo rotate at all between when the servo Enable signal goes inactive and the brake locking on.
There is a very distinct 'click' when the brake either engages or disengages, and that is an auditory confirmation that everything is working.

The relays you have linked to have a catch diode on the coil of the relay. The catch diode is the 'catch' the voltage spike that occurs when the current is removed from the coil. You do not want a several hundred
volt inductive spike going back into your controller. A regular diode will work, but they are a bit slow. Better off using a highspeed diode, or for little relays such as this a 1N4148 signal diode is any amount good enough.
I buy surface mount 1N4148's for less than 2 cents each in 1000 off lots.

What the relays do not have is any protection on the contact side. The electromagnetic brake is pronouncedly inductive. When the relay contacts open where is that current going to go? It will spike voltage
sufficient to cause an arc between the contacts.That is where you need a catch diode, so that is in addition to the one on the relay coil. In this circumstance an 1N4148 in insufficient, it is certainly fast enough but
not the current handling. This is where you need a high speed diode.

Craig

[peteeng]

Hi Craig - The current going to the brake coil is DC so is inductance a real problem? Peter