MC6 motor controllers & Pak 5

Archiver

Bill,

You mentioned before that you have 30A motors. What size wire are you using
to carry power and ground to/from them? Even .1 ohm of resistance in your
wire will drop 3V! I wonder too about noise resetting either the Stamp or
the PAK. Separate power supplies with a heavy single-point ground would be a
good idea.

Regards,

Al Williams
AWC
*Measure 8 pulses: http://www.al-williams.com/awce/pak7.htm




>

---

Original Message

---

> From: Bill Katakis [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=GZrYqrxBy9_5UwcCNhkFP2zvgM7H7BWGLwZ9JE6TQbXZc2ALXhtwB1rT8LsrSbisg4pEq_rKdz4_huFbmfwrCmg8]bkatakis@m...[/url
> Sent: Thursday, November 16, 2000 2:40 PM
> To: BASIC STAMP MAILING LIST DO NOT CC
> Subject: [noparse][[/noparse]basicstamps] MC6 motor controllers & Pak 5
>
>
> Some progress to note, on trying to control 3 12vdc motors via Al
> Williams Pak5 and Pak6 chips. Could be my pwm lines and ground were
> too long , they are about a foot long. I'm guessing at that. I have
> the pak5 on a seperate little board with a definite link to power and
> ground between the stamp2 , pak 6, and pak5a. The stamp vss also
> grounds to each motor controllers , which each use 3 stamp lines for ,
> enable, forward, and reverse. The pak5a is for sure grounded I thought
> through the stamp,, but here is the weird thing,, I added an additional
> ground jumper from the pak5a ground post to the MC6's micro ground,,
> suddenly I had some motor movement!! Being stupid I thought that if one
> ground jumper worked a little, then 2 would be better,, even better
> motor performance,, so I added a third and now have full control of the
> one motor controller that is hooked up. How long can the pwm lines
> be?? Will this grounding situation be complicated by adding more
> controllers? Whats happening? Duh
>
> Bill
>
>
>

Archiver

Some progress to note, on trying to control 3 12vdc motors via Al
Williams Pak5 and Pak6 chips. Could be my pwm lines and ground were
too long , they are about a foot long. I'm guessing at that. I have
the pak5 on a seperate little board with a definite link to power and
ground between the stamp2 , pak 6, and pak5a. The stamp vss also
grounds to each motor controllers , which each use 3 stamp lines for ,
enable, forward, and reverse. The pak5a is for sure grounded I thought
through the stamp,, but here is the weird thing,, I added an additional
ground jumper from the pak5a ground post to the MC6's micro ground,,
suddenly I had some motor movement!! Being stupid I thought that if one
ground jumper worked a little, then 2 would be better,, even better
motor performance,, so I added a third and now have full control of the
one motor controller that is hooked up. How long can the pwm lines
be?? Will this grounding situation be complicated by adding more
controllers? Whats happening? Duh

Bill

Archiver

At 12:39 PM 11/16/2000 -0800, you wrote:
>Some progress to note, on trying to control 3 12vdc motors via Al Williams
Pak5 and Pak6 chips. Could be my pwm lines and ground were too long , they
are about a foot long. I'm guessing at that.

That's a fact. Not only are they long, but they are probably acting like
antennas !

>I have the pak5 on a seperate little board with a definite link to power and
ground between the stamp2 , pak 6, and pak5a. The stamp vss also grounds to
each motor controllers , which each use 3 stamp lines for , enable, forward, and
reverse. The pak5a is for sure grounded I thought through the stamp,, but here
is the weird thing,, I added an additional ground jumper from the pak5a ground
post to the MC6's micro ground,, suddenly I had some motor movement!!

Now you have TWO of the THREE grounds REQUIRED ! One more to go ! Now take the
"micro ground" back to the battery, and you MAY have something that's marginally
operative ! Next step is shielding the control wires.

In ALL cases I'd use shielded wires, or cover each wire (each being INCLUSIVE OF
ALL) in wire braid. At .60 per foot you can't beat the price, and figuring in
the aggravation factor, no hobbyist should be without it. In my case, for the
control leads, I'm using an internal braided shield with a PVC jacket, WITH a
separate drain wire which is connected AT ONE END ONLY ! This is the same type
of wire which is often used for commercial intercom systems NOT the kind which
used to run though out your house. In addition to that, the internal layout of
the wires (the "conductor braid" or "twist") is "random lay" - meaning that the
twist varies randomly, with each foot. In your case, I'd make my own braid
(random lay) and improve the randomness to each three inches or four twists to
the foot, up to three feet, if required. In my experience, anything short of a
meter or so in length, usually requires this type of treatment. Over a meter,
the wire can generally be left as is, with the factory "twist", THEN - your
unplanned antennas are GONE !

> Being stupid I thought that if one ground jumper worked a little, then 2
would be better, even better motor performance, so I added a third and now have
full control of the one motor controller that is hooked up. How long can the
pwm lines be?? Will this grounding situation be complicated by adding more
controllers? Whats happening? Duh

Bill -

EVERYTHING should be grounded, and all control wires SHOULD be shielded. This is
why ground planes and grounding blocks were invented. All shields should be tied
together, all drain wires (if used) should be tied at ONE END only, and all Vdd
connections MUST be connected at some central point. All power neutrals should
be connected, and tied in some way to the Vdd connections by some solid method.
DON'T rely on simple crimp style pressure terminals. Use copper lugs with
anti-oxidant grease, nuts and bolts, solder, or similar techniques.

Even in mains service wiring, here in the US, there are occasions where TWO
ground paths are required by N.E.C. code. One is the (green or bare) ground
wire, and the other is a "mechanical bond wire". This is separate and distinct
from the neutral (generally white in AC) which is a separate matter altogether.

IMHO, grounding, shielding, instrumentation cabling, and like topics OUGHT to
appear on this list more often. I claim NO EXPERTISE, I just know what works !

I'm looking at a catalog right now (MPJA [noparse][[/noparse] http://www.mpja.com ]) where
Instrument cables are listed. These are Tektronix shielded data cables 20" long
configured as 8-conductor DB-9 to DB-25 and they can be had for $ 3.95. 3/8"
ground braid is .60 per/ft. and is sold by the foot. A hacked set of cheap
battery jumper cables from K-Mart or Wal-Mart works fine for the power leads,
just make sure they're copper and NOT aluminum or steel. Generally they're too
short for anything practical anyway ! If you have need for any other cabling or
grounding equipment, please contact me off list, as I have ample resources at my
disposal.

Hope that helps, as I know you've been in contact with Carl, and have discussed
this with him. These MCBs work great, once you get them set up, and part of that
is proper wiring technique. How any of this may be impacting the PAK
co-processor, is more for AL to say, then me.

Regards,

Bruce Bates

>Bill