Can I "beef up" an existing R/C ESC by adding MOSFETs?

Bite the bullit and buy a 10A ESC...

In principle, yes, you can add MOSFETs in parallel and make everything work better. In practice it's probably easier and less expensive (when you consider burnt components) to make your own ESC from scratch. It's easier and less expensive yet to just buy an ESC that'll supply full current to the motor.

If it's pulling 1.2A unloaded then it'll probably pull up to 20A during acceleration. So your ESC is only about 7 times too small.

On the bright side a cheap Chinese ESC that's up to the task shouldn't be too spendy. Check out HobbyKing, or HobbyPartz, and see.

--
www.wescottdesign.com

[...]

[...]

Hi, TTman.

Not sure what Steve McQueen has to do with this (), but you're probably right. It's just so much more appealing to try fix the consequences of one's previous ignorance by waving a magic soldering iron.

Ah, well.

Frank

--
  Perfect as the wing of a bird may be, it will never enable the
  bird to fly if unsupported by the air.  Facts are the air of
  science.  Without them a man of science can never rise.
                             -- Ivan Pavlov

SNIP

My m8s 300A ones get a bit pricey tho :)

[...]

[...]

[...]

Hi, Tim.

Thank you for the followup.

Ah. Paralleling the existing MOSFETs (assumption) didn't even occur to me, as the BB-3-9 is shrink-wrapped and undoubtedly contains surface mounted parts. No, I was thinking of something black-box-ish and much more perverse, where the existing ESC's

+/-PWM output was fed through a couple of comparators and their outputs used to drive a fresh, new H-Bridge.

OOK! (Verbalization, not a modulation reference. )

So the symptoms I'm seeing sound to you like an overloaded ESC?

One of the things bothering me was that I was seeing this "sluggishness" even when the motors weren't loaded... but I could run the motors full forward for an extended period _without_ seeing the sluggishness.

I'd looked at HobbyKing earlier today, but hadn't known about HobbyPartz. Thank you. Most of the data I need for matching up my needs _is_ available, but hard to spot; I wound up clicking on every ESC in the right current range that didn't say "brushless", but still had to squint a lot.

Here's what I'm (now) looking for:

- 20A or more (sustained) - Brushed motors - 18V battery: 15-16 "LiXX cells", or (18/3.7) = 5 LiPo cells. - Forward/Reverse - A BEC that delivered enough current for the receiver and the steering servo would be nice, but not essential.

Here's the closest I've been able to come to what I need:

$55 Pololu 18V25 (assembled): 25A, 5.5-30V, fwd/rev.

Sigh. I'll have to go looking again.

Here are my writeups on rcgroups.com, if you're curious:

Reviving a Sears C3 RC Truck (slippery slope!)

Intermittent sluggish performance from homebrew R/C Truck

Thanks again for the feedback.

Frank

--
  ...[I]t was assumed by American statesmen that whatever was uttered
  or urged in the name of moral or legal principle bore with it no
  specific responsibility on the part of him who urged it, even
  though the principle might be of questionable applicability to the
  situation at hand and the practical effects of adherence to it
  drastic and far-reaching. We were at liberty to exhort, to plead,
  to hamper, to embarrass. If others failed to hear us, we would cause
  them to appear in ungraceful postures before the eyes of world
  opinion. If, on the other hand, they gave heed to our urgings, they
  would do so at their own risk; we would not feel bound to help them
  with the resulting problems -- they were on their own.
            -- George F. Kennan / American Diplomacy 1900-1950

• To Electronics | | | | + +-------------------+ .-. | | | | | |0.25 ohm | '-' | + | | | | + | 100 | | ___ LED | +--+|___|-->|-+-----+ | | | | (Battery (+)) - --- |

At 10 amp draw from your battery, this will generate 2.5 volts into the LED that has a 100 R for the sink. It should just start to light the LED at 10 amps depending on what kind of LED you use.

From your description, it almost sounds like you have a feed back recovery error that is accumulating and you are seeing the results of this. In other words, when you change direction request, it isn't resetting the loop error back to zero and instead, it's decrementing the error and is slowing the response down on your request.

Jamie

...

...

Did it work before you started screwing around with it?

Thanks, Rich

[...]

[...]

Jamie,

Thank you!

A Visual Ammeter! Brilliant! (Actually, I was hoping the LED would stay _dark_... but it didn't.)

There were a pair of 0.47ohm 5W resistors just sitting on the original Sears PCB, giving me 0.235ohm and 10W (the I*I*R calculation says I should be using a 25W resistor, but I don't plan on having the LED lit continuously).

I built the circuit on a portion of an old RS proto PCB, and built a couple of Tamiya power connectors so I can patch it into either the ESC-Motor or the Battery-ESC circuit. The one change I plan to make (tomorrow) is soldering a second LED back-to-back with the first one so I can also see "overloads" when the motor is running in reverse.

What you're describing does fit with what I saw this evening. Following a fresh power-up, if I gradually increase forward throttle I see the LED flicker very briefly at the beginning, but once the motors are turning well the LED stays dark.

After four or five bang-bang throttle changes (full forward to full reverse and back) the motor doesn't respond as well, and the LED stays lit longer. When the problem is os bad that the motors barely turn the LED stays lit _almost_ solidly (there is some flicker). At least, that's what happens in forward; I won't be able to monitor reverse until I add that second LED.

So... now that I can reproduce the problem fairly well, tomorrow's project is to hook up my DMM and a 'scope so I can actually _see_ what the ESC is delivering when the LED lights up.

The fact that the LED lights for more than brief fractions of a second is a sign that my motors deserve a "bigger" ESC. BaneBots has a 12A unit, but after reading Tim Wescott's message I want something that can handle more current. And, in any case, I don't think I should purchase a second product from a company that won't answer its e-mail.

Thanks again for the circuit. I had all the parts lying around, but it would have taken weeks (or longer) for it to occur to me to use them in that particular way.

Frank

--
  If teachers now lack the knowledge they need to teach reading and
  other subjects well, it is not because they are innately
  incompetent but because they have been trained under faulty
  romantic ideas about the nature of reading and the worthlessness
  of "mere iformation". -- E.D. Hirsch, Jr./The Knowledge Deficit

Hi, Rich.

A fair question (), but the short answer is "Nope".

The longer answer is that one doesn't expect much from a transmitter-less battery-less $3 thrift store purchase. I'm just pleased that the motors and gear train appear to be in good shape.

Frank

--
  The thing which keeps life romantic and full of fiery possibilities
  is the existence of these great plain limitations which force all
  of use to meet the things we do not like or do not expect.
     -- G.K. Chesterton: On the Institution of the Family (1905)

[...]

Some results, for anyone who isn't thoroughly bored with this thread:

Battery pack fully charged (well, four hours' worth, anyway). Motor load limited to gearbox (truck inverted, wheels spinning freely... or maybe that's its way of asking to have its battery hatch scratched ).

First: Jamie's Visual Ammeter is hooked between the battery and the BaneBots 3.5A ESC, and my DMM isconnected to the + and on the battery side of the JVA.

1) No throttle: Vbatt is 20.4V 2) Advance the throttle slowly: neither LED lights,and after 30 seconds of full forward throttle it settles at 19.0V 3) After some bang-bang fwd/rev throttling, the motors turn "sluggish" one LED or the other lights, and Vbatt is at 19.8V.

So heavy current _is_ getting drawn when the problem occurs, but the battery pack looks good.

Second: Move the JVA to the motor side of the ESC and add my Tek465 in parallel with the DMM.

1) No throttle: Vmotor is at 0V. (This is good. ) 2) Full forward: 18.63V, full rev: -18.62V. 3) With "sluggishness" condition recreated, Vmotor is... _very_ low, varying from 1V down to 0.5V (negative if in reverse), and it stays that low util I release the throttle completely. One LED or the other is fully lit.

Here's the scope trace (10V/div and 0.5msec/div). The "PWM" pulse is about 2div long, making the PWM rate around 1kHz. (Any errors will be blamed on my keyboard.)

Light fwd throttle:

20 |\ |\ || || || || 10 || || --+|+-----------+|+--- || || 0+...|/............|/.........

Medium fwd throttle:

20 |\ |\ | \ | \ | | | | 10 --+ |+---------+ |+--- || || || || 0+.....|/............|/.......

Nearly full fwd throttle:

20 |\__ |\__ --+ \__ +----+ \__ +---- || || 10 || || || || || || 0+.........|/...........|/....

Full forward throttle is simply a straight line... as one might hope.

And, when the motors "go sluggish", the scope trace is a mere twitch ("under 1V") at the PWM pulse start point followed by a decay to zero.

Having written all this, I see "LEDs fully lit" in my notes, but... if the delivered voltage to the motors (okay, to the motors with the JVA's 0.235ohm resistor in series), where is the current coming from to light the respective forward (or reverse) LED? The slowly-turning motors decreased their resistance?

It's late and I'm tired. I'll see what things look like in the morning.

Frank

--
  ...[I]f we admit that that there must be varieties in art or
  opinion, what sense is there in thinking that there will not be
  varieties in government? The fact is very simple. Unless you are
  going deliberately to prevent a thing being good, you cannot
  prevent it being worth fighting for. If there were no longer our
  modern strife between nations, there would only be strife between
  Utopias. -- G.K. Chesterton: Mr. H.G. Wells and the Giants (1905)