Mutiplexing Opto/Fet, Circuit provided.

Thread Starter

quant

Joined May 21, 2009
114
The following circuit shows a Solenoid which is 5v signaled and 24v operated at 2.4A!
I need to cascade the Solenoids in a 100x10 = 1000 Solenoid Matrix by Shifting Register method.
So, how can take a positive line from the below circuit to my Columns of the Solenoids and the Ground to the rows of the Matrix?
Obviously there will be 110 of these available.
thank you for your help

 

Attachments

Last edited:

SgtWookie

Joined Jul 17, 2007
22,230
Unfortunately, you created the screen shot image in .jpg format, so it's impossible to read several of the component values.

It's also a bit small, making it harder to read.
Please re-create the screen shot capture, but a bit larger, and save it using .png format. MS Paint has this file/save format option.
.png format files work great for these types of images, as .png is very compact, and is not "lossy" like .jpg format files are.
 

SgtWookie

Joined Jul 17, 2007
22,230
MUCH better, thank you!

OK, how fast do you plan on turning these solenoids on and off?
What is the resistance of the solenoids? If you don't know that, what is their current draw at 48v?
[eta] Never mind - you said it's 2.4A at 24v. However, you have 48 volts across the solenoid. Did you mean to do that?

R2 is probably too low in value. 620 Ohms will get about 6.3mA current through the IR emitter.
[eta]
The IRFZ44 has a very large gate charge requirement. It would take a long time to charge it through a 10k resistor.
There is no discharge path for the gate except the leakage current through the Zener. It would take a very, very long time for the MOSFET to turn off.
 
Last edited:

Thread Starter

quant

Joined May 21, 2009
114
Hi,
Thank you for your help.
Actually the speed is not much of an issue, since the Solenoids turn on once a second for a duration of 1 second and then that particular Sol. rests for about 20 minutes!
Please explain the 48V across the Sol/, no I did not mean to do that must be my mistake.
I'm so far using 300 Ohms for R2 and will try upping it to 600 and see.
The IRFZ44 charge delay is negligble to me and I don't even notice it but it's most interesting the discharge issue you have brought up, I was wondering what the heck was going on and could not understand the phenomena...thanks a lot for mentioning it and is quite true it takes about > a couple of seconds to discharge which in fact is a desired feature for my use(saves me adding the wait feature in the program... :) ). In case I needed to get rid of it what would I needed to do?
I have been put on the Cross Point Switch track now and am deep into studying it, what do you think of that approach?

again, thanks for all your help here.
 

CDRIVE

Joined Jul 1, 2008
2,219
Hi,
Please explain the 48V across the Sol/, no I did not mean to do that must be my mistake.
Judging by this response you don't have 48V. Sgt. Wookie (I think) mistook your -24V on your print as being just that. Did you mean -common or GND?
 

SgtWookie

Joined Jul 17, 2007
22,230
Judging by this response you don't have 48V. Sgt. Wookie (I think) mistook your -24V on your print as being just that. Did you mean -common or GND?
That's the thing, I see +24v and -24v. To me, that means a difference of 48v. I have to take the schematic literally.

If the intent is +24V and GND (or 24v return), it needs to be corrected. If the 24v return is not common with ground, that needs to be indicated as well - or the 24v supply could be represented by +12v and -12v.
 

SgtWookie

Joined Jul 17, 2007
22,230
Hi,
Thank you for your help.
Actually the speed is not much of an issue, since the Solenoids turn on once a second for a duration of 1 second and then that particular Sol. rests for about 20 minutes!
OK, so the cycle time isn't a big deal.
Please explain the 48V across the Sol/, no I did not mean to do that must be my mistake.
Well, when I see +24v and -24v, that is in relation to ground; so you wind up with 48V!
I'm so far using 300 Ohms for R2 and will try upping it to 600 and see.
OK, but 620 Ohms is a standard value of resistance; 600 Ohms is not.
A decade table of standard resistance values is here: http://www.logwell.com/tech/components/resistor_values.html
Try to use E24 values (those in the green columns) - it will make things a lot easier when you actually go to buy parts.

The IRFZ44 charge delay is negligible to me and I don't even notice it but it's most interesting the discharge issue you have brought up, I was wondering what the heck was going on and could not understand the phenomena...thanks a lot for mentioning it and is quite true it takes about > a couple of seconds to discharge which in fact is a desired feature for my use(saves me adding the wait feature in the program... :) ). In case I needed to get rid of it what would I needed to do?
MOSFETs have gate capacitance; they need to be charged and discharged.
A slow gate charge/discharge time is a bad thing. When that happens, the MOSFET dissipates power as heat. This will result in your MOSFETs popping like popcorn.

I have been put on the Cross Point Switch track now and am deep into studying it, what do you think of that approach?
I have no clue what you are talking about. :confused:
 

SgtWookie

Joined Jul 17, 2007
22,230
OK, I took a liberty or two with your schematic, as I am wont to do. ;)



The MOSFET is a little SOT23 package rated for 2.7A. It'll take 22A for short bursts. Based on your 1-second ON and lots of time OFF it should be OK.

It has a tiny gate charge compared to the IRFZ44.

It will "snap" on and off a lot faster than what you had. It's still not ideal, but I'm trying to help keep your costs down.

Note the change in the optocoupler pull-up resistor, and the addition of the gate pull-down resistor.

The inductor L1 represents your solenoid; it has a resistance of 10 Ohms.
 

Attachments

Thread Starter

quant

Joined May 21, 2009
114
sgntWookie,
Thank you kindly for all your help so far,

- The decade table will help this rookie a lot and sorry about the +/-24 mix up got that too, I guess then I should have said the Com of the 24V?

- Also got the MOSFET concept, it helps to understand how it works.

- Just saw your Schem(my last post was 5am my time had to get some shut eye!), thanks in advance will comb through it and get back.

- The Cross point switches are a fantastic concept that I have just learnt about, here:

Multiplexers

........................
...........A cross-point switch is usually an M x N device, whereby any or all of M inputs may be connected to any or all of N outputs (and vice versa).

http://www.maxim-ic.com/appnotes.cfm/an_pk/638

Google - http://en.wikipedia.org/wiki/Crossbar_switch

http://www.altera.com/support/exampl...ver_16x16.html
 
Last edited:

SgtWookie

Joined Jul 17, 2007
22,230
- The decade table will help this rookie a lot
It helps a great deal to know what values of resistance you can actually purchase!
If you need an odd value of resistance, you can usually use a couple of resistors in series or parallel to get what you need. Here's a handy online calculator:
http://www.qsl.net/in3otd/parallr.html
and sorry about the +/-24 mix up got that too, I guess then I should have said the Com of the 24V?
That would work - or just refer to it as ground.
 

SgtWookie

Joined Jul 17, 2007
22,230
Is this about right please?

You have the P-ch MOSFETs upside down.
Their drains need to go to the +24v supply.

The anti-EMF diodes across the coils need to have the cathodes towards the +24v supply.

You're doing something odd with the grounding side of the coils. It's not going to work like that.
 
Top