Go Down

Topic: Do I really need R3? (Read 937 times) previous topic - next topic

DangerToMyself

Nov 02, 2019, 03:53 am Last Edit: Nov 04, 2019, 05:53 pm by DangerToMyself
I'm in the middle of designing a project using multiple stepper motors and a DC motor. I'm planning to make a PCB to hold the shift registers and make a lot of the connections.

One thing to keep in mind. No two motors will ever be running at the same time.

Below is a picture of my schematic (first time using Eagle). I know it's not perfect. Tips are welcomed!

I had this sketched out for a few months now and can't seem to recall how I came about adding the resistor R3. I'm getting better with this type of thing. But I'm still not sure that it is even necessary. Nor what size if it is needed. All I had in my notes/sketch was a resistor symbol there.

All help is appreciated.

UPDATED VERSION IN POST 12 BELOW


groundFungus

#1
Nov 02, 2019, 04:20 am Last Edit: Nov 02, 2019, 04:27 am by groundFungus
I think that R3 would be there to limit the gate current while the gate capacitance charges.  There has been debate about whether the resistor is necessary.  I always use one of around 180 Ohms on 5V systems.
You will save everyone's time if you read and follow the forum guidelines.  :)          
https://forum.arduino.cc/index.php/topic,148850.0.html
and
https://forum.arduino.cc/index.php?topic=97455.0

DangerToMyself

I think that R3 would be there to limit the gate current while the gate capacitance charges.  There has been debate about whether the resistor is necessary.  I always use one of around 180 Ohms on 5V systems.
Thank you. I'll be sure to update my breadboard layout and try it with the 180 Ohm just to be on the safe side. Not that I doubt you. I doubt me.

Off to the Eagle forums to figure out how to get rid of the "No SUPPLY for implicit POWER pin IC1P VCC" error I have. Need 5v attached to pin 16. But symbol doesn't show pin 16. Nor pin 8 which is GND. Hmmm.

Grumpy_Mike

I would also add a 10K resistor from pin 6 to ground, to prevent any glitches during power up.

DangerToMyself

#4
Nov 02, 2019, 06:35 pm Last Edit: Nov 04, 2019, 05:53 pm by DangerToMyself
I would also add a 10K resistor from pin 6 to ground, to prevent any glitches during power up.
Messed up the original somehow. As I mentioned, first time using this program. But this is the same as it was above with the 10K resistor added.

Is that where you recommend putting a 10K ?? (R4)

UPDATED VERSION IN POST 12 BELOW

Grumpy_Mike

Quote
Is that where you recommend putting a 10K ?
Yes.

krupski

#6
Nov 03, 2019, 07:21 am Last Edit: Nov 03, 2019, 07:26 am by krupski
I think that R3 would be there to limit the gate current while the gate capacitance charges.  There has been debate about whether the resistor is necessary.  I always use one of around 180 Ohms on 5V systems.
A resistor is not necessary because the Rds(on) of the port driver mosfet limits the peak current that the port can output and in the microseconds (or less) that it takes to charge (or discharge) the gate (iincluding the Miller effect), the total heating of the driver mosfets cannot even be measured.

Not to mention that every gate inside the MCU experiences momentary short circuit spikes millions of times per second as it switches (hence the need for bypass caps!) and isn't hurt one bit.

Those of you who think a resistor is needed to limit the Rds(on) current would fab a microcontroller with so many resistors that we would be back in the stone ages of simple computers that occupied an entire room and couldn't run faster than 100 kHz...

In fact, all the resistor will do in slow down the switching time of the motor driver mosfet (keeping it in the linear region longer) which will just cause it to run hotter and waste power.

Maybe a semester of EE101 would help to understand this...?
Gentlemen may prefer Blondes, but Real Men prefer Redheads!

krupski

I would also add a 10K resistor from pin 6 to ground, to prevent any glitches during power up.
Agreed with this.
Gentlemen may prefer Blondes, but Real Men prefer Redheads!

krupski

It is a MOSFET there will not be gate current. You should know how much current is reqyired at the output and then decide the value of the resistor
There is a momentary current when switching from high to low or vice-versa. It is the result of charging or discharging the mosfet gate capacitance (magnified by the Miller effect). After that, of course, the steady state gate current is zero.

Surely you understand that people who use gate resistors do so to limit the momentary current spike caused by charging or discharging the mosfet gate, not to limit the (zero) "steady state gate current".
Gentlemen may prefer Blondes, but Real Men prefer Redheads!

MarkT

#9
Nov 03, 2019, 03:17 pm Last Edit: Nov 03, 2019, 03:19 pm by MarkT
There's no ground return on connector J4.  You should never run logic signals without ground return wire(s) in the same cable.  Particularly if running clock signals.

Come to that, the stepper controller connectors don't have a ground return either.
[ I DO NOT respond to personal messages, I WILL delete them unread, use the forum please ]

DangerToMyself

There's no ground return on connector J4.  You should never run logic signals without ground return wire(s) in the same cable.  Particularly if running clock signals.

Come to that, the stepper controller connectors don't have a ground return either.
Care to elaborate? Like I said, I'm getting better, but still a long way to go. J4 is connected to ground. So, I'm not following you there.

As for the steppers, they are powered from a separate source. But I think I see what you're saying. I need to pull a wire to this board so it all has a common ground.

Hmmm. I brought the 5v to the board for the DC motor. No idea why I didn't add power to the steppers. Thinking a little editing is needed.

DangerToMyself

#11
Nov 04, 2019, 05:49 pm Last Edit: Nov 04, 2019, 05:50 pm by DangerToMyself
There's no ground return on connector J4.  You should never run logic signals without ground return wire(s) in the same cable.  Particularly if running clock signals.

Come to that, the stepper controller connectors don't have a ground return either.
I see where the issue is. You're referring to J4 in the original image I posted. I messed that file up somehow and things got renumbered differently in the later versions. Thus, I was confused.

And to add to it, they got renumbered again. Here is my latest. I think this covers everyone's concerns so far.


Grumpy_Mike

Quote
I think this covers everyone's concerns so far.
The 10K pull down should be on pin 7 not pin 6, on that circuit. It needs to be on the pin you are driving the FET from.

DangerToMyself

#13
Nov 04, 2019, 06:18 pm Last Edit: Nov 04, 2019, 06:19 pm by DangerToMyself
The 10K pull down should be on pin 7 not pin 6, on that circuit. It needs to be on the pin you are driving the FET from.
Doh!!!



fixed it

MarkT

You still don't have ground returns on every connector.
[ I DO NOT respond to personal messages, I WILL delete them unread, use the forum please ]

Go Up