Maker Pro
Maker Pro

Hall effect controlled fuel injector

Cepo

Mar 3, 2018
11
Joined
Mar 3, 2018
Messages
11
Hello. I'm new to this forum. I have very limited electronic experience.

I am looking to power a 9v or 12v circuit board that controls a fuel injector. A hall effect sensor inside the crankcase (oil and heat resistant) would be triggered from a magnet on a part of the rotating mass.

I have seen several circuit boards controlling an ignition coil. However, they are using a hall effect that wouldn't work in this application and energizing a coil that may vary a great deal of resistance from that of an injector.

I just received an injector. Unfortunately, I don't know what the coil resistance is yet.

Any help would be appreciated.

Thank you,

Mike
C.E.P.O.
 

Bluejets

Oct 5, 2014
6,901
Joined
Oct 5, 2014
Messages
6,901
Rail to rail open collector such as an Allegro A1120 and use it to drive a mosfet. Control whatever you want.
 

Alec_t

Jul 7, 2015
3,587
Joined
Jul 7, 2015
Messages
3,587
A fuel injector usually requires a high current pulse to get it to open quickly, followed by a reduction in the current to a holding level to keep the injector open for the required time without overheating its coil. There are dedicated ICs availble which facilitate this, e.g. LM1949, TPIC44L01.
 

Cepo

Mar 3, 2018
11
Joined
Mar 3, 2018
Messages
11
Rail to rail open collector such as an Allegro A1120 and use it to drive a mosfet. Control whatever you want.
I looked up the Allegro A1120. I would not be able to use that sensor in this application without some type of plastic protection. The sensor will be in the crankcase with hot oil flying about. I'm still trying to figure out what a mosfet is and whether there's a delay once it detects a signal. There is very limited time for the injector to open and close (.00105 seconds @ peak RPM). This may be able to be extended, but not by much (+ .000175) to .00123 seconds.
 

Cepo

Mar 3, 2018
11
Joined
Mar 3, 2018
Messages
11
A fuel injector usually requires a high current pulse to get it to open quickly, followed by a reduction in the current to a holding level to keep the injector open for the required time without overheating its coil. There are dedicated ICs availble which facilitate this, e.g. LM1949, TPIC44L01.
There is very limited time for the injector to open and close (.00105 seconds @ peak RPM). This may be able to be extended, but not by much (+ .000175) to .00123 seconds.

I'm not too worried about getting the pulse width dialed in at this time. I'll save that after my proof of concept is achieved. The injector will only operate after a designated RPM is reached, which would be controlled using a TPS of sorts.

That LM1949 is encouraging. I don't know where/how the control receives a signal. If it's a plug and play set up between the hall sensor and the injector that would save a lot of time and money, especially at my customer's end. I'll look more into this. Thank you, Alec!
 

Alec_t

Jul 7, 2015
3,587
Joined
Jul 7, 2015
Messages
3,587
What do you have in the way of power supplies for this project? If the injector coil has significant inductance you may need quite a high voltage to get sufficient current to open the injector in <1ms.
The LM1949 is rather long in the tooth, so may not have been designed to cope with such brief pulses. Check the datasheet for timing issues.
 

Cepo

Mar 3, 2018
11
Joined
Mar 3, 2018
Messages
11
What do you have in the way of power supplies for this project? If the injector coil has significant inductance you may need quite a high voltage to get sufficient current to open the injector in <1ms.
The LM1949 is rather long in the tooth, so may not have been designed to cope with such brief pulses. Check the datasheet for timing issues.
A 9v battery will open the injector. A 12v battery pack may be more suitable for longer life. This is going on a small engine (200cc), with RPM up to 9500 (4750 at the cam). I can place the sensor so there is lead time before the injector opens, but that lead time needs to be pretty consistent so as not to have the injector opening too soon or staying open too late. Basically, they injector needs to open right when the intake closes, inject fuel, and close before the plug fires. At 9500 RPM that's not a lot of time. Lol

I should add, although the engine rpm would be 9500 max, the injector will pulse 4750/min max (3500/min minimum at 7000 RPM)
 

Alec_t

Jul 7, 2015
3,587
Joined
Jul 7, 2015
Messages
3,587
I can place the sensor so there is lead time before the injector opens, but that lead time needs to be pretty consistent
Won't the lead time need to be changed as the rpm increases?
 

Hopup

Jul 5, 2015
253
Joined
Jul 5, 2015
Messages
253
What is the application of this kind of design If I can ask? Is it for better control over the high rpm range or something else?
 

Cepo

Mar 3, 2018
11
Joined
Mar 3, 2018
Messages
11
Won't the lead time need to be changed as the rpm increases?
Yes/maybe. That part will come once the concept proves to work. In the meantime, getting the injector process complete within the time frame at max rpm is the goal. I do also recognize the higher the rpm the less time to inject fuel (more fuel at lower rpm vs less fuel at higher).

By the way, the 'fuel' is nitrous.
 

Cepo

Mar 3, 2018
11
Joined
Mar 3, 2018
Messages
11
FB_IMG_1520100472667.jpg Here is a picture of the type of engines we build.
 

Hopup

Jul 5, 2015
253
Joined
Jul 5, 2015
Messages
253
I assume you are talking about nitromethane, that will sure give it some good power. I wonder could you just "vaporize" it using nozzle and spray it in the intake in controlled way for same result?
 

WHONOES

May 20, 2017
1,217
Joined
May 20, 2017
Messages
1,217
I thought Piezo fuel injectors were the preferred choice these days. Faster and more accurate.
 

Bluejets

Oct 5, 2014
6,901
Joined
Oct 5, 2014
Messages
6,901
Don't see why you have any sensor inside the crankcase. All relative information can be collected externally. Former only creates another unnecessary complication.
 

Cepo

Mar 3, 2018
11
Joined
Mar 3, 2018
Messages
11
I assume you are talking about nitromethane, that will sure give it some good power. I wonder could you just "vaporize" it using nozzle and spray it in the intake in controlled way for same result?
The drawback on these engines are the short port length. Port injection disturbs the aif-fuel flow and ratio. It has been done that way. It works. It's also inefficient (current systems use a continuous burst even when the intake valve is closed) and provides an unstable air-fuel mix as its injected into the port within an inch of the carb/valve.

The benefit of direct chamber injection is a continuous air-fuel ratio of gasoline and air, which is easily adjusted to the desired air-fuel ratio to suit the amount of nitrous that is allowed to enter the chamber. One of the more important aspects is increased compression. A 1cc shot of nitrous boosts these engines from 12:1 to 13:1. Then, of course, you have the boost in power from the nitrous itself.

I'm talking nitrous oxide dry shot.
 

Cepo

Mar 3, 2018
11
Joined
Mar 3, 2018
Messages
11
Don't see why you have any sensor inside the crankcase. All relative information can be collected externally. Former only creates another unnecessary complication.
Please explain how to trigger an injector from something outside these engines.

Add: These are 4 cycle engines.
 

Cepo

Mar 3, 2018
11
Joined
Mar 3, 2018
Messages
11
I thought Piezo fuel injectors were the preferred choice these days. Faster and more accurate.
I did see piezo injectors just the other day. However, i don't know if they can be installed for direct chamber injection (not port injection). I will follow up on this.
 

Bluejets

Oct 5, 2014
6,901
Joined
Oct 5, 2014
Messages
6,901
Please explain how to trigger an injector from something outside these engines.

Add: These are 4 cycle engines.

Yes, I am very conversant with engines and technology.
I build both here for many years.

Certainly.

Use a crank position sensor and a cam position sensor.

Feed relevant data into a uC, fire the injector control at appropriate time and, depending on the nitro content, either hold on to your hat or get a fresh set of undies.

While you are at it, if not already done so, perhaps fit through-bolts to hold that head on.
Threading into Al casing will, sooner or later, end up with lift-off.
Plastic tubing would be considered risky at minimum.
 
Last edited:

Cepo

Mar 3, 2018
11
Joined
Mar 3, 2018
Messages
11
Certainly.

Use a crank position sensor and a cam position sensor.

Feed relevant data into a uC, fire the injector control at appropriate time and, depending on the nitro content, either hold on to your hat or get a fresh set of undies.
The cam is located inside the engine. There is no part of the cam outside of the engine.

The bolts are not an issue in pulling out. Sealing the head can be problematic without o-ringing or treating the head to reduce flex.

The plastic tubing is for venting the block and providing direct oiling to the valve train.
 
Last edited:
Top