Controller for 4 wheels (2 dual motor drivers / speed controllers)

[swabygw]

My bot has 4 wheels/gearmotors (Pololu: 50:1 Metal Gearmotor 37Dx54L mm with 64 CPR Encoder
).  I'm hoping to find a programmable controller that can control forward/reverse/speed of all 4 wheels independently OR controls both left/right.  The best programmable controllers I've seen available have dual motor-drivers and I don't want to have to buy two controllers or do an H-bridge.  Any suggestions for a programmable controller that has either two dual motor-drivers or can drive all 4 motors?  Thanks, in advance.

P.S., If I use a dual motor-driver, is it possible to wire the left wheels to one driver, the right wheels to the other driver, since the wheels on each side will always be working together?

[waltr]

If I use a dual motor-driver, is it possible to wire the left wheels to one driver, the right wheels to the other driver, since the wheels on each side will always be working together?

Yes, if the maximum current draw for both motors does not exceed what the driver can handle.

[desirez]

Hi all,

 I would like to revive this thread as I have similar wishes. I need to control 4 wheels, in a "tank" like situation. Whereby using skid steering.

Is there a "quad" motor driver or such out there? I have no worries with programming so ideally, the cheapest one/pair going?!

Any help appreciated

[newInRobotics]

It depends on how much current motors You are willing to use consume. If they are small motors, then somethng like a pair of L298 should do.

[desirez]

Hi Newinrobotics,

 The motor I would like to use is:

http://docs-europe.electrocomponents.com/webdocs/0bf3/0900766b80bf315d.pdf

No load current 0.9A, Max eff current, 5.5A..

[desirez]

These look like a great solution, however the price is very high. If I need one per motor on a 4wd system then thats £200.. would it be possible to pair sides as mentioned above?!

http://www.dimensionengineering.com/SyRen25.htm

[newInRobotics]

No load current 0.9A, Max eff current, 5.5A..

No then, 5.5A is waaay too much for L298   Something like 2x Sabertooth2x10 would do better, although it gets expensive.

You can always attempt to build Your own four channel motor driver made of MOSFETs  

[newInRobotics]

If I need one per motor on a 4wd system then thats £200.. would it be possible to pair sides as mentioned above?!

You can use one Sabertooth dual 25A motor driver and run two motors per channel, however, it is not advisable as same side motors will have slightly different speeds causing higher current consumptions and drift.

[newInRobotics]

FYI: With something like 2x IRF 9530 P-CHANNEL MOSFETs and 2 x RFP12N10L N-ch MOSFETs, few resistors, few diodes, and 2x smaller N-ch MOSFETs You can build 1 H-bridge to drive 1 motor. Circuit would look like the one underneath:


Click on the image to open simulation

[rbtying]

Since Black Friday is coming up--Pololu has a number of nice motor drivers which are ~50% off--see here: http://www.pololu.com/blackfriday.

[desirez]

Thanks for all the responses guys. Does the H bridge allow for regenerative power capture? I was under the impression it did not, unless this power can be diverted away from the bridge?

As an aside, are cordless drills using regenerative capture or presumably its a cost to worthwhile for the potential?! (looking through a document that suggests using drill motors/circuitry as an economical alternative).

Appreciate you guys know your stuff, this is still somewhat of a learning experience for me so apologise for my naivety

[bens]

Hello.

I just want to point out that it's pretty much pointless to use four motor drivers in your application since the two motors on each side should always be turning at approximately the same speed (you don't gain anything by having them turn at different speeds or in different directions, especially if they are linked by tank tracks).  I recommend you use one motor driver per side (or a dual motor driver), and connect the two motors on each side in parallel to a single motor channel.  You could use two of our Simple Motor Controllers for this (they can be easily paired for mixed RC control, or you can chain them together for control with a single serial line), or could use something like our dual VNH5019 motor driver shield (with or without an Arduino).  Both are part of Pololu's upcoming Black Friday sale, as are our metal gearmotors.

- Ben

[bens]

Thanks for all the responses guys. Does the H bridge allow for regenerative power capture? I was under the impression it did not, unless this power can be diverted away from the bridge?

The H-bridge circuit in general allows for regenerative braking, but it is hard to do well and it depends on all kinds of variables, many of which are specific to the actual motor you are using, so I'm skeptical that the controllers out there that claim to have regenerative braking actually do all that much.  Ultimately, I think the effect will not be all that significant, and I personally do not think it should should be an especially big consideration when you choose a controller.

- Ben

[desirez]

Thank you for your reply Ben, some thought provoking stuff. Perhaps then, I can do away with regen, in that the extra battery capacity is more of a reliable source for extra power.

For now, I have all my answers Thank you once again! Bring on black friday sales!

[Soeren]

Hi,

Regenerative breaking (RB) pops up every now and then, but I have to agree with bens on its lack of importance in hobby robotics.

I would imagine that it would be something very popular and P.C. in current school projects, as they typically are more P.C. and pop than really thought through.

So, if it earns you any points, do go for it but don't imagine it's the least bit "green".

First, how large small a percentage of the run-time do you need as fairly hard a breaking as would give any usable amounts of regen. energy?
Second, consider the charge efficiency of around 65% tops of battery cells - and how high low is the efficiency of your homebrew regen. circuitry?
Third, what's the environmental footprint of the components going into the regen circuit?

That's just the main factors, that makes me believe that a hobby robot will actually be less green using RB (and you can quote me on that in your report, whether or not you go with RB ).

It's good thinking about the environment, but it's bad jumping in blindfolded, to go with the flow without considering where it takes you.

[desirez]

An interesting angle. I guess for me the idea was that under no applied current to the motors (ie slowing down under force loss) - with the hbridge correctly opened, whatever speed I was slowing down (faster obviously under RB) would only go to top up battery voltage. Perhaps its ignorant of me to assume the Hbridge can be induced to allow current both ways (since it takes very little voltage to switch the mosfets - losses would be at the charge side of around 0.65 as you say).

D.

[newInRobotics]

Hi desirez

To charge battery You need to control voltage and current in a certain manner, it is not wise to just apply increasing voltage in the inductor straight to the battery, You most definitely fry something.

[Soeren]

Hi,

An interesting angle. I guess for me the idea was that under no applied current to the motors (ie slowing down under force loss) - with the hbridge correctly opened, whatever speed I was slowing down (faster obviously under RB) would only go to top up battery voltage. Perhaps its ignorant of me to assume the Hbridge can be induced to allow current both ways (since it takes very little voltage to switch the mosfets - losses would be at the charge side of around 0.65 as you say).

It can be done, but the net gain is extremely marginal - like perhaps 30 seconds on a charger (just guessing, as I have no chance of knowing how often and how long you would be braking,biut the more you brake, the more power you have to spend on acceleration).

The 65% I mentioned is solely the charge (in)efficiency of the battery, mainly due to losses in the chemical process (this is why most battery chemisties heat up during charging). The other losses in the complete system is lowering this quite a bit further.

But, as mentioned, if it's for educational merit, go for it anyway - you can always argue that it isn't very efficient to get extra credit