Need help with solenoid curciut

[GRC]

Hi all.
I'm currently finshing a project which uses a PIC 16F628
to control a solenoid. The problem is turning the solenoid
on and off. The solenoid in question is in my son's "paint-
ball-marker". It needs to turn on and off rapidly. The solenoid
has a resistence of about 1.2 Ohms, and uses 9VDC. My
guess is this will draw 7.5 amps. The output is opto-islolated'
but the transistor doesn't seem to supply enough current or
voltage (part # NTE 196 npn transistor).
If I place the leads of the solenoid across the + and - of
the 9VDC, the solenoid pulls in and operates correctly,
but the power source of my regulated 5VDC fails and
causes my PIC to reset (believe this is a problem with
captacitance).

If anyone has any suggestions or ideas I would greatly
appreciate them.
Thanks For Reading.
Gordon

 

[Oznog]

Well, you've got a bit of a problem there. The battery does not have enough current capacity to trigger than relay. Lemme guess, this is a 9V battery? That can put out something like 300ma, 500ma or so.

Best thing is to use a logic level n-channel MOSFET to trigger the solenoid. Why? Well, this figures into the plan...

The reg generally won't drain power off the 5V rail if the source voltage falls. So if we can put a decent amount of capacitance on the 5V side and keep the power usage to a minimum, the 5v rail can survive a momentary power loss. An npn transistor driving 7 amps requires hundreds of times more base current than the entire system would require with a MOSFET. So if you watch power consumption from sensors, etc and keep the PIC's freq as the minimum needed to do the job, you can keep the 5V rail away from brownout for several moments.

You must- MUST- have a flyback diode the relay's coil.

It's still a huge problem that there's just not enough power on the battery, it's just not cricket to load a battery so high. At best, you could use a huge capacitance. We're talking huge, if you want to have a cap that supplies 7 amps for 1/10th sec with only a 2v drop this would need to be 350,000uf. Now there are "supercapacitors" which can do this, they can be in the range of 100,000uF to 1F. But they're usually only made as 5.5V ratings at most. You could potentially put 2 in series (the total capacitance is halved), but some protective measures must be taken to assure the center terminal does not float away from 9v/2 or it will blow one of the caps.

There is also the issue of the internal resistance of the cap. Most are not made for charge/discharges in the range of amps, their terminal voltage will drop a lot and it can damage the cap.

 

[Russlk]

If you are deriving the 5 volts from the 9 volts, then the problem is that the 9 volts supply drops below the minimum input for the 5 v regulator. If the 9 v and 5 v supplies are seperate, I don't understand why there is a problem.

 

[GRC]

Thanks for your replies.
What I'm attempting, is to replace the electronics
that are already in the paint-ball-marker. The original
electronics operated off a nine volt battery, used a
PIC 16C505 and a 6800uf cap rated for 16v. All of
this fit into the handle.

I know how to write the code, and control inputs and
small load outputs, but I'm stuck on how to put the
curcuit together for such a big load. Will this curcuit
still require the diode across the solenoid when its
opto-isolated (I'll put one in to make sure)?
I'll replace the NPN transistor with the N-channel
Mosfet you suggested. Perhaps a second battery
dedicated for the 5VDC with its ground tied into
the 9VDC ground will solve the power failing problem.
What size caps are needed for the 5VDC?

As for frequency, the 16F628 has an internal oscallaltor,
which I believe runs at about 4Mhz. Will this frequency
draw a large amount of power?

One last note/question, the solenoid is only on for a few milli-
seconds at a time. Will "Inductance Reactence" play a factor
in overall current draw to the solenoid? How phiscally big
are the super caps you mentioned. Small enough to put in
the handle?



Thanks For Your Ideas and Solutions
Gordon

 

[Russlk]

Yes, you need the diode but don't need the opto-isolator if you are operating from batteries. The inductance of a solonoid is usually low and the series resistance is high enuf to limit the current, so the time constant will be short. You need to look up the data sheets for the 5 v regulator and micro, that will tell you all you need to know.

 

[gerty]

What is the battery life on this thing? Are your mods going to extend the life or just shoot faster?

 

[GRC]

Thanks again for the your replies.
I will download the data pdf files (good idea thanks).
As for battery life, dont know if this will extend
the duration of charge. The goal is make the
paint ball gun have 3 different modes of operation,
single fire, burst fire and automatic. Every thing
is good to go with the execption of the solenoid
curcuit. The cool thing about to program is that
the timing for firing speed can be adjusted in
program. This device had a bad habit of the paint
balls getting stuck half way down the drop shute
and wedged against the plunger. This was caused
from firing too fast, faster than gravity could drop
a new paint ball into the drop shute.


Thats probally more than you really wanted
to know. Hope this gives a better idea of what
I'm attempting.

Any more Ideas or Suggetions are welcomed.
Thanks Again Everybody
Gordon