Maker Pro
Maker Pro

Current limiting strategies for L298N?...

BGB

Nov 30, 2014
154
Joined
Nov 30, 2014
Messages
154
I somehow thought you were using a Sumtor 42H34913A4 stepper motor, but can't find that anywhere on this thread. What stepper motor are you using for the XYZ axis? What stepper motor are you using for the spindle drive?

yes, that is the correct motor.

info I have on the motors:
detent torque: 0.40 Nm
step torque: 1.0 Nm
rated current: 1.3 A
step: 1.8 degrees (200 steps per revolution)
rated RPM: 600 RPM (from Chinese sites)
voltage: 24v (also from Chinese sites)
(measured) resistance: approx 3 ohm IIRC.

cost was: $16 per motor

I am running them at 12v, and thus far haven't got them going quite so fast, but this may be due to voltage.


the spindle motor is currently a normal Brushed DC motor.

spindle motor: ADRS550SH
voltage: 9-16v
max RPM: 20,000
efficiency current: 10A
stall current: 40A
stall torque: 0.34 Nm (IIRC)
(measured) no-load current: around 2A
(measured) resistance: approx 0.33 ohm

cost: $10 per motor

also have another type of motor available: CRS-550PH-8022AF
found stats were fairly similar to the motor above.
(measured) no-load current: around 1.4A
(measured) resistance: approx 0.2 ohm
(observed) slightly quieter than the above motor.

cost: $11 per motor

(actually, had intended to get more of the prior motor, but the seller sent me these motors instead).

the spindle is being driven directly by PDM, and will use a hand-wired H-bridge (will build one specifically for this, have used prior H-bridges with the same type of motor though, but they have more limited power output).

for the H-bridge, I will probably either use some MOSFETs (IRF540), or maybe doubled-up MJE3055T transistors (*), which should be able to handle around 20 amps or so...

*: would have to double-up the drive transistors and then do multi-stage Darlington boosting, which would end up requiring a fair number of transistors (12x MJE3055T, 4x 2N2222, and some 2N3904's as well). may be better just to use the MOSFETs.
 

BGB

Nov 30, 2014
154
Joined
Nov 30, 2014
Messages
154
FYI here's a good application note that describes various types of motor drivers: https://www.silabs.com/Support Documents/TechnicalDocs/An191.pdf

yeah, nifty.

though some of it does seem to assume a specific type of controller (for example, in their C code, and some examples assume the ability to use analog IO and similar).

otherwise, still waiting for diodes to show up... (so I can build the stepper driver boards).

also now have some OSB available (cheaper than plywood), so may be able to build an initial version of the box for the thing.


in the meantime, had been working on the tachometer for the tool motor.
discovered that the phototransistors I have are a bit slow and problematic (they could still work, but are unlikely to be useful much above around 9000 RPM at one pulse per revolution, forcing them to be used for monitoring the spindle, and needing to isolate the spindles' reflection band from the background IR reflection).

tried making a magnetic pickup coil, but the output voltage was very low (around 1mV) and I was unable to successfully amplify the signal (I suspect wire was too short and too thick for this).

had a bit more success building some magnetic switches, where the field from a spinning magnet vibrates a small ball of twist-tie wire on the end of a piece of springy wire, causing it to endlessly open and close contact with a blob of solder (but it is annoyingly delicate and takes a bit of "fine tuning" with hot-glue to make it work). I mostly "tuned" them using headphones to monitor their output. I guess normal people would use a hall-effect sensor or similar here (don't have any around, would need to order some).

have a few other ideas which I may test (namely, using a sliding contact with nail-polish as an insulator), ... which could potentially be less problematic than the use of a glued-on magnet. also half wondering if "motor noise" can be used somehow (will infer that, given no one does so, the signal is probably too much "polluted" to be all that useful for measuring RPM).
 
Last edited:
Top