Which transistor should i use?

Thread Starter

Thomas0zero

Joined Dec 31, 2011
6
I need some circuitry help for something i have in the making.

I need a switch that will be receiving a pulsing signal and i need it to switch every time it receives a signal. For every 2 signals i only want the destination to receive one.

NO SIGNAL
SIGNAL (Sends current through switch then the switch goes to the switch off position)
NO SIGNAL
SIGNAL (signal is blocked but now the switch goes to the on position)

repeat

It also has to switch quickly. The signal will be under 12v.
 

Adjuster

Joined Dec 26, 2010
2,148
This sounds like an application for something like a CMOS divide by 2 flip-flop, rather than discrete transistors, but this is not really my sort of thing.

A fuller description would help others to comment, for instance, exactly how quickly does this thing need to switch? How long are the pulses in time, and how often do they occur. How much output current is required?
 

Thread Starter

Thomas0zero

Joined Dec 31, 2011
6
I need the output voltage to stay the same.

To be more specific im trying to interrupt a fuel injector for a vehicle.

So i need very fast switching times.

I was told i needed a transistor, but something else might get me my end result.

Like i said, i need the signal to do the switching as well.
 

SgtWookie

Joined Jul 17, 2007
22,230
Hello Thomas0zero,
Sounds like you might be attempting either something like the Northstar/Aurora system's "limp-home" mode, or something like the old Cadillac 8/6/4 system.

Unfortunately, you've run right into one of the restrictions of the Terms of Service on this website; the relevant portion of which reads (excerpted and highlighted for your viewing pleasure):

6. Restricted topics. The following topics are regularly raised however are considered "off-topic" at all times and will result in Your thread being closed without question:
  • Any kind of over-unity devices and systems
  • Automotive modifications
  • Devices designed to electrocute or shock another person
  • LEDs to mains
  • Phone jammers
  • Rail guns and high-energy projectile devices
  • Transformer-less power supplies
Even though making a "limp home" mode might be "a good idea", you probably will run afoul of emissions controls laws. Turning off injectors every other cycle will indeed allow the engine to run cooler, but your ECU won't know that it is supposed to be happening; and it will try to increase the fuel flow because of the presence of oxygen in the exhaust.

So, due to the Terms of Service, and even though it sounded like a good idea at first, the thread is going to have to end here.

Hopefully, you'll understand that the rules were established for very good reasons, and by the owners of the site - and they are not subject to negotiation - and accept the rules gracefully.

This isn't an attempt to "cheat" you out of your project - it's simply that we don't support discussions of this nature.

If you feel that you must pursue the project, there is a short list of member-selected forums over in the "Electronic Resources" forum where you might inquire; here is a direct link to that thread:
http://forum.allaboutcircuits.com/showthread.php?t=54400

In the meantime, if you have other projects that don't conflict with the ToS, why not start a new topic/thread, and we'll see what we can do to help you with it/them.
 
Last edited:

Thread Starter

Thomas0zero

Joined Dec 31, 2011
6
Ah. well its not acctually a automovitve modification.

Im trying to build an Injector cleaning station for my shop.:D

It will be something similar to this so i can test flow rates as well.

 

SgtWookie

Joined Jul 17, 2007
22,230
Ahh, well that IS different.

What kind of frequency range are you talking about?

A 4-stroke engine will fire a given injector once every other rotation, so an engine running at 7,200 RPM will fire a cylinders' injector 60 times/second.
7,200/2 = 3,600; 3,600/60 = 60.

Were you planning on firing them a lot faster than that?

What voltage range are you going to be using?

Injectors used to fire at around 3.3v, but I've been hearing of some pretty high-voltage injectors lately.
 

Thread Starter

Thomas0zero

Joined Dec 31, 2011
6
You know, im really not sure what voltage to fire them at just yet. Ive been meaning to get a multimeter after my car but i figured i would get the circuity designed first.

You know now that you mention it i would like to be able to demonstrate how quickly injectors fire. So hopefully we can use parts that can switch that fast.

I work up at Chrysler and this could be a good teaching tool or demonstration for customers, "why to clean your injectors."

I would have not need to go faster than 7,200.

Any ideas?
 

Adjuster

Joined Dec 26, 2010
2,148
You should find out as much as possible about what your circuit has to do, including the output levels, before starting to design it.

A multimeter is something you should get sooner rather than later if you want to do anything with electricity, but this cannot tell the full story about your injectors, which require short pulses, and quite high currents with the current rapidly changing during each pulse.

Proper measurement of fuel injector currents would mean using an oscilloscope with a current probe. This may be beyond what many of us would be able to do at home. See this article - but note that the injectors referred to may be different from your ones. http://www.atsnm.com/images/escope/fuelingectortutorial/FuelInjectorWaveforms.htm

It would be better to get information about injector currents and voltages from their manufacturers.
 

Thread Starter

Thomas0zero

Joined Dec 31, 2011
6
i didnt think it would be that big of a deal since i wouldnt be altering the power output, just leaching off it to switch the switch.

Man, i really thought this would be a simple question. lol
 

Adjuster

Joined Dec 26, 2010
2,148
i didnt think it would be that big of a deal since i wouldnt be altering the power output, just leaching off it to switch the switch.

Man, i really thought this would be a simple question. lol
Building a circuit to test injectors is by no means impossible: it has been done before, and you might even be able to find a suitable circuit if you like on the internet. Here is an example: but I make no warranty that it is correct or would suit your needs. http://www.dinofab.com/fuel_injector_tester.html

Nevertheless, I repeat what I said about first finding out the requirements of the type of injector you have: they are not all the same, for instance diesel injectors typically require massively more power than petrol (gasoline) ones.

As to your (lol) comment, quite possibly the injectors will be just fine with just a switch connecting them directly to the 12V input, as in the example circuit. Lots of people on this forum know about electronics, but are not necessarily automobile experts: that is your department, but earlier you said that you are not sure of what voltage is required.

This is potentially a fairly simple project, but the current levels are high enough to take it out of the trivial category.
 

SgtWookie

Joined Jul 17, 2007
22,230
Here is an example: but I make no warranty that it is correct or would suit your needs. http://www.dinofab.com/fuel_injector_tester.html
DON'T build either of those! The ON-time duty cycle to the injector is greater than 50%, and the current will only be limited by the MOSFET Rds(on), the injector itself, and the wiring! I'm afraid that Dino is not quite aware of what he's subjecting his components to.

In both schematics, on the right-hand 555 timers, the timing components are the 100 Ohm resistor (call this R1), the 470 Ohm resistor (call this R2) and the 22uF capacitor (call this C1).

The ON time is a function of (R1+R2)*C1*.69
The OFF time is a function of R2*C1*.69
As you can see according to the formula for this configuration, the ON time must always be greater than the OFF time.

There is another more subtle problem, and that is the low value of R1. C1 is discharged via R2 by pin 7, which is an open-collector input; it connects that point to ground when the cap is supposed to discharge. With a 12v supply, current through R1 will be 12v/100 Ohms = 120mA, or 0.12 Amperes! This is just too much to ask of the 555 timer, and it is bound to cause problems.

My standard "rule of thumb" for values of R1 is a minimum of 100 Ohms per volt of Vcc; or in this case, 12*100 = 1.2k Ohms - and preferably higher.

Changing these resistors can be compensated for by decreasing the size of C1. So, if we change C1 from 22uF to 1uF, and multiply the resistors by 22, we get 2.2k for R1 and 10,340 for R2. 10k Ohms would be close enough.

The frequency is not bad @ 70 Hz. However, this change in resistors does not change the duty cycle.

Adding a 1N4148 diode in parallel with R2, anode towards R1, allows us to achieve duty cycles less than 50%. Then, changing R1 to 2k and R2 to 20k gives us right at 70Hz, with a much shorter duty cycle; this combination will give just over 10%.

To reduce the duty cycle further, C1 would need to be decreased to 0.1uF/100nF. Then R2 could be increased to 220k to keep the frequency almost the same, but just about 1% ON time.

Another problem is that the Zener across the injector is rated for 12v, but Dino is showing it being used in a car. The Zener should be a few volts higher than the unit is being powered by, to avoid excessive power dissipation in the Zener.
 
Last edited:

Adjuster

Joined Dec 26, 2010
2,148
@SgtWookie: Thank you for pointing out the error of my ways, and "Dino's". I have to admit not having made much effort to check the circuit over, which was remiss of me.

As to the drive conditions for the injector, I was somewhat influenced by the OPs assertion that these are simply switched across the supply, given that he claimed to be in the trade. I should have known better.
 

SgtWookie

Joined Jul 17, 2007
22,230
Adjuster,
You weren't in error - it's Dino's circuits that have problems.

Our old '88 Jaguar had injectors that were powered with 3.3v. I've heard someone say recently that they'd worked on a BMW that used >100V on the injectors, which surprised me - but that would certainly get them moving quickly.

All I can say about it is, one would have to go at it on a case-by-case basis.

Dino left his E-mail addy on that page; I might just send him a note and let him know about what I'd found.

Since my last post, I added a pot (VR1) so that the duty cycle can be varied from ~1%-~10%, but the frequency stays about the same. R2 was lowered to compensate for the pot.

I've attached a schematic of my mods to just the right-hand portion of the circuit.
 

Attachments

Top