Sabertooth 2x10 Problem

[rosolam]

I recently purchased a Sabertooth 2x10 and had the last two days to play with it. I found it very easy to plug and play and I loved the simple serial interface and the no solder connections.

The short story of this post though is that I was experiencing an intermittent issue yesterday (day 1) and it is dead today (day 2). I am awfully disappointed because the intermittent issue was tolerable enough but it died just as my bot was going to take its maiden voyage to test my PID/encoder logic on real terrain… what a downer!

Below is a more detailed description that I sent to Dimension Engineering last night about the intermittent issue and then a second email that I sent this afternoon about it being dead. Hoping some of the pros out there can take a quick read and give me their thoughts.

I just received a Sabertooth 2x10 that I purchased
last week. I am very impressed with it, however I am
experiencing a peculiar intermittent issue that I
can’t crack and am concerned that the item may be
defective. I have detailed the issue below and am
hoping that you can help advise me.

While operating, every couple of minutes Motor-2 will
kick into high speed and the Error LED lights up on
the board and then a moment later it drops back down
and the Error light goes out. I do not experience
this with Motor-1.

Connections:

1. A fully charged 12v Pb battery is connected to the
   Sabertooth power connections (B+/B-) carefully minding
   the polarity

2. Two new 12v gearhead motors are connected; One to
   each of the two motor channels (M1A/M1B and M2A/M2B)

3. Per the manual, the S1 terminal is connected to an I/O pin of my
   Basic Stamp II which is on a Homework Board (which
   includes an in-line 220ohm resistor)

4. Per the manual, the 0V terminal is grounded to the Vss on the Basic
   Stamp II Homework Board (which includes an integrated
   5v regulated power supply)

5. The remaining S2 and 5V are left disconnected

6. Being used in Mode 3: Simplified Serial Mode, I
   have had the dip switches set, per the manual, as
   on-off-on-on-on-on for 2400 baud as well as
   on-off-on-off-on-on for 9600 baud.

After a number of test programs to reproduce the
issue, I finally was able to get it to happen in this
simple program:

SpdCtrlPin = MCU pin connected to S1 on the
Sabertooth
SpdCtrlBaud = serial baud rate. I have tried both
2400bps (BS2 value of 396) and 9600bps (BS2 value of
84)

DO
SEROUT SpdCtrlPin, SpdCtrlBaud, [66]
PAUSE 250
SEROUT SpdCtrlPin, SpdCtrlBaud, [191]
PAUSE 250
LOOP

The issue appears to occur intermittently when
supplying a near STOPPED signal between motor A and
motor B. I have tried several variations and this is
the only one that appears to cause the issue, but
being apparently intermittent it is hard to say with
certainty.

Please advise,
Michael Rosola

Unfortunately, I had to follow up that last email with this one this afternoon:

I have likely not given you time to even read my first
email yet, but I have an unfortunate update.

This afternoon I attempted to try and further narrow
down the erratic behaviour I explained in my last
email when the motor controller suddenly became
unresponsive. I disconnected the power and found the
battery leads hot to the touch. I reconnected very
briefly and reconfirmed that there was no response, no
LEDs and hot leads. I then attempted using a small 9v
battery, again no response.

It appears that the motor controller is now dead.

I believe that I have followed all the documentation and have handled
everything properly including paying special attention
to battery polarity.

Please advise.

Thanks,
Michael Rosola

Anyone else experience this issue? Although I believe that I carfully followed the manual and everything was well within the tolerances of the device, does it appear to anyone that I did anything to cause this?

Thanks,
Michael

[robotdude]

@rosolam

Anyone else experience this issue? Although I believe that I carfully followed the manual and everything was well within the tolerances of the device, does it appear to anyone that I did anything to cause this?

Thanks,
Michael

I have been using one for a long time. No problemo.

I wouldn’t get too excited till the weekend is over. Give them a chance to read the email. I’m sure they will want to see the board so they can try and figure out what went wrong.

[beave]

I noticed similar things as you did with just one side acting erratically. Two things that I was doing wrong were causing it though I think. I haven’t had enough time to play with it yet.

1. I didn’t have the ground from the Sabertooth connected to Vss on the BS2.

2. I was using a for loop with the PULSOUT command and didn’t have it sending frequent enough signals.

The other thing I noticed was that if I turn the power on on the Sabertooth before the BS2 is on, it sometimes freaks out.

I’m using DIP settings:

Pin 1 2 3 4 5 6
0 1 1 0 1 0

(non-lithium mode, microcontroller R/C input, independent mode with linear control)

I’m using this on the 4WD3.

I’ll try out the SER commands though and see if I get the same results as you.

[dimrobot]

Hi Guys,

John from Dimension Engineering here. I designed Sabertooth 2x10 as well as the rest of our motor drivers, so I’ll be happy to help with questions/quirks/problems involving.

Michael:

You should have recieved your repaired Sabertooth 2x10 today. If so, have you gotten a chance to try it out yet? I don’t think it was anything you did wrong, I think it managed to slip through our testing with a dry solder joint.

Beave:

The easiest way to tell if you’re having hardware problems or code problems when running in independent mode is switch the inputs and switch the motors. If the motor output that’s acting erratically stays the same (always M1 or always M2) then the problem may be hardware related. If the glitching moves with the inputs, then code is probably to blame.

Its important to have the grounds connected together. Otherwise, you can’t be assured of a reliable signal.

In microcontroller mode, the refresh timing isn’t critical (you can leave it running arbitrarily long after a pulse) but it is very important to make sure that you don’t have any spurious signals on the inputs, because it won’t necessarily reject them.

If you’re powering the Sabertooth before the BS2, then it’ll be looking for signals when the BS2 starts up. Check to make sure that the lines start up as inputs or at least in a known state.

Hope that helps. I’ll keep an eye on this thread if someone needs assistance (or contact us directly by phone or email)

Thanks,
John Bellinger
Dimension Engineering

[beave]

Thanks so much for the detailed help! I actually did what you said to figure out my code problem. I switched the cables around and found that it was just a code problem. I got confused for a while but feel comfortable with it all now. There were a lot of options in setting it up which made it harder but it’s great because I can always use that controller for some application even if I don’t use it on a rover.

I haven’t tried the SEROUT command yet but did play with the PULSOUT a little more and found that it’s working great. I also put some LEDs in line with the input so I can see the actual pulses.

That’s good to know about the power up sequence. I’ll make sure to code it so the pins startup right.

Thanks again and great work on an awesome product!

[adam-21]

@DimRobot

I’ll keep an eye on this thread if someone needs assistance (or contact us directly by phone or email)

Thanks,
John Bellinger
Dimension Engineering

Hello John,

I have a question about your Sabertooth 2x5. It says the low end nominal voltage is 6V, but I was wondering if it would be ok to run it with a 4 cell NiMH pack. The motors I want to use it with are rated for 4.5-5.5v, and I would not want to push them beyond their limits because they cost about $25 each.

Thank you.

[adam-21]

Actually, nevermind about my question. I did some more research, and it looks like my little motors can handle the extra voltage. I’ll run a 5 cell pack.