Searching \ for '[EE] logic component connection limit' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=logic+component
Search entire site for: 'logic component connection limit'.

Exact match. Not showing close matches.
PICList Thread
'[EE] logic component connection limit'
2007\01\02@130541 by Andre Abelian

flavicon
face

Hi to all engineers and happy new year.

I have to drive 128 small motor thru small bridge driver. each motor needs
2 wires and I only have 8 pins available. all I have to do is select the motor
and update the direction. I plane to use 74hc373 that it can latch while
I am selecting different pair. In this case I am going to need 32 latches
connected to the PIC 8 pin bus and thru chip select update the status.
My question is can I connect 32 logic directly to pic port or I
may need to connect thru buffer.
sorry I forgot to mention that I have keypad and lcd connected to other ports.
I am using 18F4520
any idea or recommendation will appreciate.

Andre

2007\01\02@132156 by Harold Hallikainen

face
flavicon
face
Back in the good ol' days of TTL, we had fan-in and fan-out
specifications. Now, the current requirements of a CMOS input are
extremely low, and the PIC can source or sink 25mA (when operating at 5V).
The limitations now are on how long it takes to get the line up to 5V or
down to 0V while driving several chips. I don't think you'll have any
problem driving 32 chips. If you have any trouble, it may just require the
addition of a nop in your code to let the data lines settle before sending
your strobe pulse.

Good luck!

Harold


{Quote hidden}

> -

2007\01\02@132607 by David VanHorn

picon face
The limits you're looking for are called "fan-out", but it's not that simple
anymore.

HC inputs are essentially capacitors, so you need to sum the total
capacitance of traces and gates, and then look at the output drive, and
decide if the rise/fall times that result are something you can live with.

For what you're talking about, it sounds like you'll have no problem as long
as you remember that with 32 HC gates hung on your output pin, your outputs
are going to need a little time to become valid, between each state
change.

2007\01\02@160300 by Jinx

face picon face

> The limitations now are on how long it takes to get the line up to
> 5V or down to 0V while driving several chips. I don't think you'll
> have any problem driving 32 chips. If you have any trouble, it may
> just require the addition of a nop in your code to let the data lines
> settle before sending your strobe pulse

I had terrible r-m-w problems driving several 74HC gates due to
the collected capacitance of inputs and wiring. As well as the nops
found a 1k loading resistor for the PIC pin helped

This is for inputs but you could use 4017s for motor selection

http://home.clear.net.nz/pages/joecolquitt/0pots.html

As I mention the above, alternative method for multiple inputs

http://home.clear.net.nz/pages/joecolquitt/mixer.html

2007\01\02@162901 by David VanHorn

picon face
>
>
> I had terrible r-m-w problems driving several 74HC gates due to
> the collected capacitance of inputs and wiring. As well as the nops
> found a 1k loading resistor for the PIC pin helped


Yup, though RMW isn't a problem I've had to deal with in AVR land. :)

Driving a significant bus isn't trivial, and as you mention, some series
resistance at the driving end helps, by lowering the current into the bus
impedance.  Termination impedances also help, but you'll need to look at it
with a scope to see what you're doing.

I had a problem once with an Acer motherboard in a server machine due to
improper termination on a line.  The thing was giving all sorts of problems,
but once we found the line, we were able to change the terminating impedance
to get it under control.  AFAIK, acer put the fix into production on that
board.

2007\01\02@165624 by Wouter van Ooijen

face picon face
> I had terrible r-m-w problems driving several 74HC gates due to
> the collected capacitance of inputs and wiring. As well as the nops
> found a 1k loading resistor for the PIC pin helped

If you have RMW problems there is only one real cure: use a shadow
register.

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2007\01\02@174355 by Gerhard Fiedler

picon face
Wouter van Ooijen wrote:

>> I had terrible r-m-w problems driving several 74HC gates due to
>> the collected capacitance of inputs and wiring. As well as the nops
>> found a 1k loading resistor for the PIC pin helped
>
> If you have RMW problems there is only one real cure: use a shadow
> register.

FWIW, the OP says he is using the 18F4520. Shouldn't have RMW problems at
all...

Gerhard

2007\01\02@182658 by Andre Abelian

flavicon
face
Wouter,

I like to know what is RMW? and shadow register?

thanks in advance

Andre




-----Original Message-----
From: spam_OUTpiclist-bouncesTakeThisOuTspammit.edu [.....piclist-bouncesKILLspamspam@spam@mit.edu]On Behalf
Of Gerhard Fiedler
Sent: Tuesday, January 02, 2007 2:43 PM
To: piclistspamKILLspammit.edu
Subject: Re: [EE] logic component connection limit


Wouter van Ooijen wrote:

>> I had terrible r-m-w problems driving several 74HC gates due to
>> the collected capacitance of inputs and wiring. As well as the nops
>> found a 1k loading resistor for the PIC pin helped
>
> If you have RMW problems there is only one real cure: use a shadow
> register.

FWIW, the OP says he is using the 18F4520. Shouldn't have RMW problems at
all...

Gerhard

2007\01\02@184759 by Harold Hallikainen

face
flavicon
face
RMW is Read-Modify-Write. The bit set and bit clear instructions work by
reading the whole port, modifying the bit you are setting or clearing,
then writing the whole port back out again. The RMW problem is a problem
when you do a few bit sets or clear immediately after writing to a port.
Due to capacitive loading on the port pins, it takes time for the line to
go high or low. If you wrote it high, then read it before it goes high,
modify another bit, then write the modified value back out again, that
line that you had written high but had not yet had a chance to go high
will now be low (since that's how it was read). Shadow registers are bytes
of RAM where you do the bit set and clear instructions on this byte of
RAM, then copy the RAM to the port. You end up doing the RMW on the RAM,
which changes very quickly, unlike a port line. As someone else has
pointed out, if you are using an 18 series part, you can avoid the RMW
problem by doing all your outputs (such as bit set and clear) to the LAT
register instead of the PORT register. The RMW of the LAT reads the output
latch instead of the output pin, so it doesn't matter if the pin has not
yet changed to the correct level.

Hope I made sense!

Harold


{Quote hidden}

> {Original Message removed}

2007\01\02@191435 by Jinx

face picon face

> I like to know what is RMW? and shadow register?

A graphical example of rmw. An SX pin driving a CMOS 4000

http://sxlist.com/techref/scenix/sxrmw.htm

Back in the days when I thought, nah, rmw will never get me .....

The solution here was to use either a shadow register or extra
nops, and it worked perfectly thereafter. Amazing how such a
simple and, at the time, elusive, problem can be so frustrating

2007\01\02@235533 by Vasile Surducan

face picon face
Define:

1. "small motors" with voltage, current, type of motor, torque and/or
load at the shaft
2. distance between logic board and driver board, distance between
driver board and motor
3. 2 wires means "left"&"right" or "direction"&"speed" ?

My two cent opinion: forget any parallel approach and do it serially
with a  standardised bus.

Vasile

On 1/2/07, Andre Abelian <.....aabelianKILLspamspam.....mason-electric.com> wrote:
{Quote hidden}

> -

2007\01\03@032627 by Wouter van Ooijen

face picon face
> > If you have RMW problems there is only one real cure: use a shadow
> > register.
>
> FWIW, the OP says he is using the 18F4520. Shouldn't have RMW
> problems at
> all...

OOps, that is of course the *other* cure: use 18F and use it correctly
(LATx).

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2007\01\03@102750 by Andre Abelian

flavicon
face
Hi to all,

I like to thank you all for replying to my question
you guys are the best.

Andre




-----Original Message-----
From: EraseMEpiclist-bouncesspam_OUTspamTakeThisOuTmit.edu [piclist-bouncesspamspam_OUTmit.edu]On Behalf
Of Wouter van Ooijen
Sent: Wednesday, January 03, 2007 12:28 AM
To: 'Microcontroller discussion list - Public.'
Subject: RE: [EE] logic component connection limit


> > If you have RMW problems there is only one real cure: use a shadow
> > register.
>
> FWIW, the OP says he is using the 18F4520. Shouldn't have RMW
> problems at
> all...

OOps, that is of course the *other* cure: use 18F and use it correctly
(LATx).

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2007\01\03@164730 by Andre Abelian

flavicon
face
Vasile,

excellent idea let me think about it.

thanks

Andre




-----Original Message-----
From: @spam@piclist-bouncesKILLspamspammit.edu [KILLspampiclist-bouncesKILLspamspammit.edu]On Behalf
Of Vasile Surducan
Sent: Tuesday, January 02, 2007 8:56 PM
To: Microcontroller discussion list - Public.
Subject: Re: [EE] logic component connection limit


Define:

1. "small motors" with voltage, current, type of motor, torque and/or
load at the shaft
2. distance between logic board and driver board, distance between
driver board and motor
3. 2 wires means "left"&"right" or "direction"&"speed" ?

My two cent opinion: forget any parallel approach and do it serially
with a  standardised bus.

Vasile

On 1/2/07, Andre Abelian <RemoveMEaabelianTakeThisOuTspammason-electric.com> wrote:
{Quote hidden}

> -

More... (looser matching)
- Last day of these posts
- In 2007 , 2008 only
- Today
- New search...