arduino motor controllers in parallel?

hello i asked a question before about using rc car esc's instead of motor drivers. all was working well until i discovered that the brake function in the esc was interfering when i was using the servo library to control the esc when i have the object avoidance code running. so i thought that would it be possible to wire two arduino motor controllers in parallel to increase current output?

here are the motor drivers:

http://www.ebay.com.au/itm/L298N-DC-Stepper-Motor-Dual-H-Bridge-Drive-Controller-Board-Module-for-Arduino-/190779229320?pt=AU_B_I_Electrical_Test_Equipment&hash=item2c6b540488&_uhb=1#ht_5248wt_1037

L298 can be wired in parallel. No problem.

I doubt that would be enough for an RC motor though - what are the motors?

they are 25mm geared motors they pull 50ma no load @12v. max efficiency current is 350ma and stall current is 1.7A. its going to be 4wd

Chagrin:
L298 can be wired in parallel. No problem.

I've heard of people "stacking" the L293 to get greater current capabilities, but not the L298. I'm not saying it can't be done, but there isn't any mention in the datasheet for it (nor is there for the L293, for that matter).

Now - you /can/ run the L298 in "bridged mode" - which is a support mode of the L298 to allow you to run a single motor, sourcing up to 4 amps of current to it (of course, you need one helluva heatsink to do this!). Whether that board the OP linked to supports this mode natively or if you need to "hack it" somewhat, I don't know. A very careful reading of the L298 datasheet and review of the board PCB and schematic would be required to know whether it can be done or not.

But putting multiple L298 devices in "parallel" - I don't know how well this would work. As I said, I've heard of people stacking the L293 - up to three units, in fact. I don't know how it could work for long term. Both the L293 and L298 use bipolar transistor outputs, and you are not supposed to try to parallel such devices. They don't work like MOSFETs do, which can be paralleled for greater current sourcing capability. You run into one or the other transistor trying to take the load completely itself, and things just go downhill from there, with magic smoke being released.

I'd just been studying the same problem with the L293 when I took a break and saw his question about the L298 so I knew it could be done/has been done. Digging a bit deeper it looks like the recommendation is that the chips come from the same batch so they'll have the same switching times and share the current more evenly.

Nothing will blow up but you should watch the heat of the chips to make sure they're sharing well.