#robotics Logs

Aug 18 2021

05:10 AM Steffann is now known as Steffanx
12:22 PM akinode: hi, anyone here with knowledge about how to build circuits who can help with this problem in this project? https://www.reddit.com/r/ElectricalEngineering/comments/p6cu71/building_a_wired_remote_control_car_u...
01:41 PM mrdata: akinode, yes motors are inductors
01:41 PM mrdata: and they resist changes in current flow
01:42 PM mrdata: consequently there are usually diodes across DC motors
01:43 PM mrdata: to allow flow back from negative to positive terminal
01:46 PM * mrdata checks the motor spec...
01:46 PM mrdata: akinode, do you need these motors to be bidirectional?
01:47 PM mrdata: if so, the controller ends up being an H-bridge
01:48 PM mrdata: controlling crazy current spikes is fun
01:49 PM mrdata: ok the DRV8833 is a dual H-bridge, good
01:51 PM mrdata: but this isnt a stepper motor
01:51 PM mrdata: so the schematic given in the data sheet doesnt apply
01:53 PM mrdata: figure 6 showing forward and reverse drive may be tho
01:57 PM mrdata: i wonder if a TVS diode would help
02:00 PM mrdata: but there is dead time built in to the DRV8833
02:01 PM mrdata: section 9.1 seems relevant as well
02:01 PM mrdata: what is the pwm frequency you use?
02:07 PM mrdata: knowing pwm frequency and motor current should allow to calculate the appropriate value for the bulk capacitor
02:11 PM mrdata: ok the DRV8833 says typical 50 kHz
02:14 PM mrdata: so this is strange; the 3300 uF cap should be plenty
02:15 PM mrdata: but the leads to the cap are quite long in your photo
02:16 PM mrdata: maybe parasitic inductance is getting in the way?
02:17 PM mrdata: if so, add another, smaller cap, say 1 or 2 uF much closer to the board
02:18 PM mrdata: what is tghe length of the wires from the board to the motors?
02:18 PM mrdata: maybe try shortening those as well
02:20 PM mrdata: and just maybe snub RC across the motor
03:04 PM akinode: @mrdata thanks for the response, I will try to shorten the distance to the motor driver
03:05 PM akinode: the pwm frequency is set to default I believe
03:06 PM mrdata: ok
03:08 PM akinode: what kind of diode would you use here if the motors needed to turn in both directions?
03:13 PM mrdata: that's the tricky part. it would be a bidirectional TVS if it is useful at all; but i am not sure
03:13 PM mrdata: i think what youre seeing is,
03:14 PM mrdata: in section 7.3.5.3 undervoltage lockout
03:14 PM mrdata: so having a low ESR cap as close to the VM pin as possible could help
03:15 PM mrdata: 1 to 2 uF perhaps
03:32 PM akinode: the lowest I could find was 4.7uF and it's a 450V capacitor. Would that make sense for my 3V circuit?
03:32 PM mrdata: um,
03:33 PM mrdata: what is the ESR of that
03:33 PM mrdata: a ceramic cap is usually appropriate here
03:34 PM mrdata: one with low piezo activity
03:34 PM akinode: something like this? https://de.aliexpress.com/item/4000395699194.html?spm=a2g0o.productlist.0.0.bf797558041iJn&algo_pvid=400e8344-7a05-4415-828b-ded69c5a1dcf&algo_exp_id=400e8344-7a05-4415-828b-ded69c5a1dcf-40&pdp_ext_f=%7B%22sku_id%22%3A%2210000001623050076%22%7D
03:35 PM akinode: and you would use that alongside my 6V 3300uF capacitor?
03:37 PM mrdata: parallel yeah; the 450V rating is overkill
03:37 PM mrdata: but the value 4.7uF is fine i think
03:38 PM mrdata: it's the equivalent series resistance that you need to know
03:38 PM mrdata: and piezo properties maybe
03:52 PM mrdata: one or two of these in parallel might be just right, https://www.digikey.ca/en/products/detail/tdk-corporation/FG18X5R1E225KRT00/7326659
09:19 PM -!- #robotics mode set to +o by sodium.libera.chat