SparkFun Forums

[raygeeknyc]

Hi all
I'm trying to use an SN754410E to provide bidirectional control of two small DC motors. The motors have noise-elimination capacitors. My application provides 5VDC from a voltage regulator and "raw" power from a ~6VDC battery pack.

I've hooked up and tested each side of the SN754410 and am seeing the same behavior on each, so I am pretty sure that it is me. I've tried two different ICs so I don't think that it is a defective/fried chip.

Summary: all 4 motor outputs always read Vin from pin 8

Here is my circuit at a point where I am trying to test with just one motor:
Pins 1,9,16 - +5VDC
Pins 3,6 - Output to motor 1
Pins 4,5.12,13 - Ground
Pin 8 - Vin from the battery pack

By my understanding, with pins 2 and 7 LOW there shouldn't be any current on 3,6 - but when I check them with my meter, they both have Vin
So, I connected Pin 2 to +5VDC - same thing, both outputs show Vin
For completeness, I connected Pin 7 to +5VDC - same thing.

Thanks in advance,
---Raymond

[riden]

From your description, everything looks like it is wired correctly. I would double check that all four ground pins are really connected to GND, and that you have the chip wired properly (i.e., not connected backwards).

[ribbotson]

When you say pins 2 and 7 are low. Do you hold them low ? They may float high with no connection to ground.

[raygeeknyc]

Thanks for the replies
- I am pretty certain it isn't backwards - Pin 1 is counterclockwise of the notch

<fixed width ascii art>
1 U 16
2 15
3 14
...
8 o 9
</fixed width ascii art>

Holding the pins low... I didn't realize that.
So...let's see....
...

YAY!
When I connect 2 or 7 to ground, the voltage on the corresponding output pin drops to 0.54 VDC.

Thanks both of you!
---Raymond

[raygeeknyc]

Well I have the 2 motor output pins reading either vIn or a small +VDC (around .5V) - but even after reading the datasheets for the SN754410 and the L293 (including figure 5, page 9 on the L293) I confess that I'm at a loss as to how to hook these up to the motor.

Regarding figure 5, I don't see why the motor pins are hooked up to vIn as well as to the motor output. I'm also a bit ignorant as to the SES5001 - it's a diode but is specifically required or could I substitute a different rectifier diode with sufficiently high ratings?

Can anyone explain how they've hooked the motor leads, or point me at or supply a good explanation of the above (figure 5) or other applicable app note?

Thanks
---Raymond

[ribbotson]

The motor goes between the two motor output pins (3 and 6). This allows the motor to be off (both high or low), forward (3 low, 6 high), or reverse (6 high 3 low). Speed or torque are adjusted by the duty cycle of the control signals.

Your motor should have two terminals. Some have 3, but the the 3rd is usually a ground connection of the case. This can be connected to ground and a couple of small capacitors connected to the motor terminals to give some electrical noise suppression. Alternatively a small capacitor (10nF) in parallel with the motor will help.

The 4 diodes around the motor in the figure are for protection of the L293. The motor is inductive, so when the driver switches of there is a reverse inductive pulse generated. The diodes clamp this to the supply rails to protect the L293. They only conduct when the rails are exceeded.

The L293D version has these diodes on chip, the others do not including the 754410. This makes some confusion in schematics for different version. You need the diodes when driving inductive loads like motors.

For diodes they should be fast acting, but do not need high voltage. I would suggest 1N5817 or UF4001 as possible alternatives.

[raygeeknyc]

Thanks for the explanation.

I've now followed the example (figure 5 in the L293 datasheet/app notes) and used appropriate diodes. It all works, if I pull pins 2|7 and 10|14 low, I can drive the two motors birectionally.

Now, on to controlling this from my Arduino stamp!

---Raymond

[Michael J]

Hi all,

I've used the 754410 for some time to control 2 motors bidirectionally. In the past, I always tied 2 PWM lines to the 2 enable lines on the chip. I never used diodes to account for the inductive kick back because I (mistakenly?) assumed they were internal to the 754410.

Now looking at the postings here and the L293/L293d data sheet I see the diodes are present in the example circuits.

Here's my confusion. In my version of the L293 datasheet, figure 5 shows bidirectional motor control, for one motor, using the 1a and 2a pins to control the direction. I know if 1a and 2a are both high or both low, the motor will brake and stop relatively quickly.

With the PWM and WITHOUT the diodes, the motor would coast during the PWM off cycle.

Now it appears, if I connect the diodes, when the enable line goes LOW (as it would during the PWM off cycle), the motor will brake (called "fast motor stop") in the L293 data sheet.

So how then is it possible to bidirectionally drive a motor with PWM, with diodes, without braking during the PWM off cycle?

Newly confused after years of happy 754410 use,
Michael

[riden]

Sorry about not answering your question about the braking, but the 754410 does have the internal diodes. The L293D's do not. So you're not completely confused.

[ribbotson]

I guess we are all confused now !

The data sheet for the 293D says "WITH DIODES", so why do say not ?

For the 754410, I referenced this
http://www.arduino.cc/cgi-bin/yabb2/YaB ... 0888334/30

There are different diodes for different purposes.

[riden]

I'm a dunce. Sorry, should have gone to bed earlier.

According to the TI datasheets, both the SN754410 and L293D have the internal protection diodes. The L293 does not and would require external diodes for protecting the chip.

[Michael J]

Boy, I'm really sorry this issue won't just go away...

I looked at the 754410 datasheet and I saw what looked like internal diodes shown on the output pins. (As background, I've never used external diodes with this chip and have never damaged one.) Unfortunately, the same data sheet shows external diodes in the sample circuit driving a 2 phase motor. The sheet does not show a simple DC motor being driven bi-directionally. For that I went to the '293 data sheet. There, the 293 (not the '293D) had external diodes in the sample circuit. This would be OK (We'd all go out and buy a bunch of diodes!) but when I looked at the 293 truth table for a the bidirectional circuit, it looked something like this:

ENABLE 1A 2A Comment
1 1 1 Fast Motor Stop
1 0 1 Forward
1 0 1 Backward
1 0 0 Fast Motor Stop
0 X X Fast Motor Stop

In the past, I have always set 1A and 2A to both high or low whenever I wanted to brake (because setting both 1A and 2A to the same value shorts the motor terminals). Now, looking at the truth table, it appears, when the enable line is low (in the bidirectional reference circuit) the motor will brake. This would seem to be a problem because most of us drive the enable line with PWM and we don't want to brake during the low portion of the cycle.

To try to make some sense out of this, I manually checked a motor connected to a 754410. With the enable line low, I was able to turn the motor by hand (which I can't really do when the terminals are shorted).

Summarizing, protection diodes should not allow for toggling the PWM line, but we all do it so something more is going on. At the same time, the internal Darlington's should probably be damaged if the diodes are missing and yet most of us haven't destroyed any of these chips yet.

Michael

P.S. A quick search of the Web will lead to a lot more confusion. My personal view now is the 754410 is equivalent to the 293 not the 293d.