Hi Jesse,
We use the Galil DMC-4080 extensively at Oak Ridge HFIR and SNS on many beamlines. It has replaced (and continuing to replace) the Parker 6K based systems. I think we only use it with 2-phase stepper motors and mostly use the D4140 amplifier
(up to 3A), but I believe there’s a few of the newer D3547 amplifiers which can drive up to 6A.
The D4140 is 1/64 microstepping, so 12,800 steps/rev, which is fixed. The current limiting is quite coarse and just has 4 levels (.5, 1, 2 and 3A), but that’s generally sufficient for our NEMA17-34 sized motors. A few of the bigger motors
might use external drives.
The D3547 is 1/256 microstepping, so 51,200 steps/rev, again fixed. Same coarse current settings (.75, 1.5, 3.0, 6.0A). I don’t have much experience with this amplifier, but I think the coarse current settings are a bit unfortunate if you
need to fine tune something.
I can confirm the Galil works great with standard quadrature encoders and BiSS-C encoders. The BiSS-C supports needs a special order code, but the BiSS-C version works with either standard A/B quadrature or BiSS encoders.
I don’t think there’s closed-loop support for stepper motors. We usually do that via the motor record and enable retries. Although, as I write this I see that they have a ‘Closed Loop Stepper Firmware’ available.
The built-in amplifiers are grouped together in groups of 4 axes. So there’s two drive amplifier boards in an 8-axis controller. This means the axes in a group of 4 need to be of the same type (either all steppers or all servos). The controller
ELO (electronic lockout), which you might use as part of an E-stop circuit, controls output power on all 8 axes, but the controller logic power stays up. We usually build motion cabinets with separate logic and drive power supplies.
Occasionally we have seen a bad drive output on a single axis, which means you have to replace the whole controller if there’s not a spare axis to use.
The controller has a 100Mbit ethernet port for communication and a RS232 port. If you choose to use the ethernet (which we do) then be prepared to place these controllers behind a firewall of some kind as they don’t survive long on a general
purpose network that may have broadcast traffic on it.
One other thing to consider is that these are very mature products and have been around since circa 2010. I believe they are developing an ARM-based replacement, but I’ve not seen any news about it on their website. They do have a couple
of newer generation EtherCAT based products. However, in general the Galil is very reliable and easy-to-use.
Cheers,
Matt
From: Tech-talk <tech-talk-bounces at aps.anl.gov>
On Behalf Of Jesse Hopkins via Tech-talk
Sent: Tuesday, August 23, 2022 4:31 PM
To: EPICS Tech Talk <tech-talk at aps.anl.gov>
Subject: [EXTERNAL] Recommended amplifiers/drivers for GALIL motor controllers?
Hi folks,
We're currently in the planning process of upgrading a reasonable number of the motor controllers at our beamline (~60 channels worth of motors, both stepper and servo). One option we're strongly considering are
the GALIL DMC-4080 and DMC-4183 controllers. I know these have robust EPICS support, and are used by a number of people in the community.
I was hoping folks would be willing to chime in with what amplifiers and drivers you're using with these controllers, particularly if they're compatible with our use cases (see below).
Our applications are pretty simple. We have ~30 stepper motors, the majority of them don't have encoders and are run open loop, but ~10 of which are encoded and I'd like to run them closed loop. The stepper motors
are a combination of two phase bipolar and unipolar, but the unipolar ones are 6 lead and could be rewired to run as bipolar if needed. Voltage and current requirements are ~2.5-6 V and ~0.5-2 A depending on the motor (and unipolar vs. bipolar drive mode).
Microstepping is necessary, but doesn't have to be really fine (around 1/8 to 1/16 is okay). When present, we have either a/b quadrature or BiSS encoders.
We have ~30 brushed servo motors, which need to be run with a linear current drive mode, rather than PWM. They're all 24 V, with current ranges from ~0.1-1.4 A (mostly on the lower end). They're all encoded with
a/b quadrature encoders.
I have looked at the GALIL internal amplifier and drive options, particularly the D3140, D4040, D4140, and D3547. The stepper drives seem like they'd probably be fine, but the servo drive doesn't quite deliver the
voltage desired (or, for a couple of our motors the desired current). If anyone has particular feedback on these, that would also be appreciated.
We're currently using Step-Pak SPD-6B and SPD-6U drivers for the steppers, and Delta Tau Amp 1 amplifiers for the servos. At the very least, because of the limited microstepping, we're looking to move away from
the Step-Pak drivers. I would consider using the Delta Tau Amp 1 amplifiers with the GALIL controllers, but I haven't looked into how feasible that is.
----
Jesse Hopkins, PhD
Deputy Director
BioCAT, Sector 18
Advanced Photon Source