Emergency shutdown method

[TimSR2]

Set it up like a detroit 2 stroke with an air shutdown valve.

It's the easiest  surefire way to kill a runaway diesel engine.  Programmable engine controller could be used to operate air shutdown on overspeed, or overheat, or fail of rpm to cease on fuel shutdown.  Back when 2 stroke detroits were still in highway trucks this was mandatory!  They had to have both fuel and air shutdowns.

[BruceM]

I know many knowledgeable folks are married to cutting off intake air, but I think if you're automating, that adding decompression control is a better use of your time and effort, since you can then use it for automatic/remote starting, too.  This was the approach taken by Lister for the SOM, so I'm not all alone in thinking this is a good way to go.

My 6/1 has a pneumatic rack closer and exhaust valve lifter; they are operated off of the same air solenoid, much like to the SOM.  Both could be done with RC servos (giant scale) for those with electric controls. I've seen a tiny commercial servo control board that turns and RC servo into an on/off solenoid. 

BruceM
Metro 6/1- for AC power and air compressor
Off grid east of Concho AZ, at 5600 feet elevation

[carbon-rod]

Hey Bruce

Do you have a pic of your setup? I would actually like to look into decompression automation as well, for the exact reason of being able to start with a smaller starter motor, either that or just put an oversized starter on there... I think I might eventually have to use a proxy to pick up spokes for secondary speed control rather than just relying on AC frequency capturing

[dieselgman]

Bruce, decompressor control for remote starting is fine and indeed what the SOM controls are about... however using that for shutdown could become an issue for possible valve collision with piston crown - especially after carbon has had the chance to build up on the piston (thereby reducing clearances). Lister did not recommend using the decompressor for shutdown. None the less, this could be done (especially for an emergency control) - but with very careful adjustment and measurement to ensure ample clearances are always maintained.

dieselgman

[BruceM]

I found some old photos of the exhaust valve lifter.

Since you mentioned it, I included a picture of the Cherry gear tooth sensor I use for speed monitoring, with it mounted next to the flywheel spokes.

I used pneumatics for my Listeroid controls since it is also my air compressor, and I have a Gast 4am based starter with a McMaster rubber friction wheel (another air cylinder presses it against the flywheel).

Dieselgman is right to point out the importance that your valve clearance is sufficient. Since the
SOM valve lifter was likewise directly connected to the rack closer, using decompression during shut down was in fact not an operational problem for the SOM, and that has been my experience as well (about 2000 hrs). I could easily separate the rack and valve lifter by adding another solenoid valve, and I might do that just to save wear on the lifter, someday.

I was careful to check my valve clearance- and set up my lifter so that it does not lift the valve quite as far as the manual "valve holder".  I have sufficient clearance to allow for carbon buildup. 

It's my experience that the pushrods of the Listeroids are soft enough to flex if you do something foolish on assembly after decarboning and don't have valve clearance.

[carbon-rod]

ahh nice.... did you get the starter idea from bendy2000(youtube)? He runs a similar setup...

that's certainly nice and simple, I have never looked into pneumatics but they seem like a great idea for mechanical actuation, possibly cheaper than electrical and hydraulic?

[Lloyd]

mmm this wouldn't work though if the rack was stuck?

I am all for simple, however I am kind of defeating the purpose by building a control system and adding electronics for it any way, but that's ok for my purpose as it doesn't need to be sitting in a field where it is required to be very reliable, for me I would rather a system that is fully automated, and because I have an electrical background that is actually half the fun automating thing is what I like to do..

You are over thinking this thing...use energize to run..

Lloyd

[Lloyd]

I found some old photos of the exhaust valve lifter.

Since you mentioned it, I included a picture of the Cherry gear tooth sensor I use for speed monitoring, with it mounted next to the flywheel spokes.

I used pneumatics for my Listeroid controls since it is also my air compressor, and I have a Gast 4am based starter with a McMaster rubber friction wheel (another air cylinder presses it against the flywheel).

Dieselgman is right to point out the importance that your valve clearance is sufficient. Since the
SOM valve lifter was likewise directly connected to the rack closer, using decompression during shut down was in fact not an operational problem for the SOM, and that has been my experience as well (about 2000 hrs). I could easily separate the rack and valve lifter by adding another solenoid valve, and I might do that just to save wear on the lifter, someday.

I was careful to check my valve clearance- and set up my lifter so that it does not lift the valve quite as far as the manual "valve holder".  I have sufficient clearance to allow for carbon buildup. 

It's my experience that the pushrods of the Listeroids are soft enough to flex if you do something foolish on assembly after decarboning and don't have valve clearance.

BruceM, 

good to see ya back.

L

[carbon-rod]

You are over thinking this thing...use energize to run..

Lloyd

The system would need to be directly connected to the spring loaded flap with the solenoid running continuously in order to open the flap. Basically it needs to be fail safe so that if the control circuit craps itself it will just shut down automatically and not require 12v to be injected by a control circuit, also if the power cable broke to the solenoid while it wasn't activated when you went to try and activate it for an emergency shutdown it wouldn't work.

[BruceM]

The idea of a rubber roller friction drive starter was something I got from Utterpower, and George B. would be the first to tell you that there is nothing new about that.

I used the Gast motor instead of an auto electric starter motor for drive as I already had an air powered woodshop and some motors left over.  This was about 7 years ago, and I did actively publicize the use of the Gast 4am/2am as a simple starter with McMaster rubber friction drive on the Listeroid forums of the time. I was pleased that others also found this easy to make and effective. They used a variety of homebrew drive rollers and enhanced it by using a single air supply solenoid, with small regulator to adjust the engagement force of the actuator. I didn't expect that an air starter would be popular, as I thought I was one of the few with a ready air supply.

Air cylinder actuators are cheap and very powerful, and can be easily and cheaply plumbed with 1/4" OD drip tubing and fittings .  The only downsides I've found with air controls are:

1. Things like regulators and check valves, and the decompression valves on my compressor head all do leak slowly to some extent. (I switched to a motorized ball valve to shut off air when the Listeroid is not in use, and use a manual ball valve to close off my  reserve air tank. My main air storage tank is 500 gallons.

2. Operationally you must be vigilant regarding keeping condensation water out of the air motors as the cast iron body will corrode, and the motor will start binding. The Listeroid's need for manual lubing every 8 hours of run time is fortunately compatible with with the need for bleeding off water from the air system, when lots of air compressing is done.  I only have to be vigilant during our monsoon season.

[quinnf]

Bruce, good to see you back.  Hope all is going well.

FWIW, you were my inspiration for the control system for my 6/1 that I'm slowlly implementing at my wife's place in Washington.  Especially as regards the decompressor and fuel rack, which are both controlled by pneumatic cylinders retracting against spring pressure when air pressure is present.  Any fault in the air system, or if something goes wrong will automaticallyl close the fuel rack and actuate the decompressor cam by spring pressure.  Can't get very much more KISS than to rely on a spring to shut off compression in a diesel.

I'm using a Toyota Camry starter turning a 2.5" rubber roller (Mcmaster) to drive the flywheel, held in place by a 2" air cylinder with spring retract.  Two air solenoids and 3 air cylinders will control the whole system.  

Air cylinders are cheap and reliable and simple to implement.  Has KISS written all over it.  www.surpluscenter.com is a great source for cheap air cylinders.  McMaster.com has them too http://www.mcmaster.com/#standard-air-cylinders/=foutwv, and the hardware (clevises, pushrod ends, etc.) you might need to mount them.  It's a natural with air starting like Bruce has done, but even though I'm using an electric starter on my 6/1 I'm actuating the starter and the shutdown paraphernalia via air because it's cheap and easy and fail-safe. 

Quinn

[Lloyd]

You are over thinking this thing...use energize to run..

Lloyd

The system would need to be directly connected to the spring loaded flap with the solenoid running continuously in order to open the flap. Basically it needs to be fail safe so that if the control circuit craps itself it will just shut down automatically and not require 12v to be injected by a control circuit, also if the power cable broke to the solenoid while it wasn't activated when you went to try and activate it for an emergency shutdown it wouldn't work.

On an energize to run system, everything has to be working else the system won't start. So it goes like this, the fuel rack solenoid only opens when in the run position, now the over temp, and low oil are also wired in this system, along with over/under speed, then so is the butterfly valve on the intake(you can get one of these from Granger) the butterfly valve can run on the same linkage to the run solenoid for the fuel rack.

So a fault at any system monitor ie over/under speed, low oil, high temp, de-energizes the run-solenoid, which then closes the fuel rack, and the intake butterfly valve, if you want belt and suspenders I would use a separate solenoid for the butterfly valve. I would also put a temp sensor on the exhaust.

Lloyd

[rcavictim]

You are over thinking this thing...use energize to run..

Lloyd

The system would need to be directly connected to the spring loaded flap with the solenoid running continuously in order to open the flap. Basically it needs to be fail safe so that if the control circuit craps itself it will just shut down automatically and not require 12v to be injected by a control circuit, also if the power cable broke to the solenoid while it wasn't activated when you went to try and activate it for an emergency shutdown it wouldn't work.

On an energize to run system, everything has to be working else the system won't start. So it goes like this, the fuel rack solenoid only opens when in the run position, now the over temp, and low oil are also wired in this system, along with over/under speed, then so is the butterfly valve on the intake(you can get one of these from Granger) the butterfly valve can run on the same linkage to the run solenoid for the fuel rack.

So a fault at any system monitor ie over/under speed, low oil, high temp, de-energizes the run-solenoid, which then closes the fuel rack, and the intake butterfly valve, if you want belt and suspenders I would use a separate solenoid for the butterfly valve. I would also put a temp sensor on the exhaust.

Lloyd

If you want a nice belt buckle for those suspenders Lloyd one could add an optical opacity gauge (LED optocoupler + some simple comparator electronics) to shut down if too much grey/black smoke in the exhaust has crossed a nastiness threshold.

[BruceM]

I like the sound of your system, Quinn.  By using an electric starter, you now need very little air, your plumbing, regulator and air solenoids can be the smallest and cheapest ones, and your air budget is essentially only for leaks.  I think your failsafe scheme is excellent.

The safety/failsafe issue is a good one to ponder for a while.  My highest design priority was to maintain "manual reversion"; the ability to operate the Listeroid completely manually with no power, no battery, no AVR (harmonic only), with no special operations to do so. This was in part because with so much of my own, new, prototype hardware and software, I wanted the piece of mind of knowing that I could at immediately operate manually "no matter what". I needed the Listeroid power for compressed air daily during 3 years of construction (even my skill saw is air) and could not go without.  I also had the unusual need to run my engine with just the air compressor, no AC at all, since I can't be near the spinning generator head for more than 30 seconds without it affecting my epilepsy.

After that, I decided that since I had no wife or kids involved, that I would just opt. for the simplest (for me) solutions. If anything fails on my system, the engine just keeps running; there is no watchdog timer on the processor or any other failsafes.  The controller does monitor for low air pressure, and will shut down for that, low oil, vibration, high temperature, no load timeout (great for people like me with memory problems) and under/overspeed. If air pressure is suddenly lost (all 500 gallons of it?!), then I'm just back to the manual operation that most guys have.  Likewise with the loss of the controller or it's 12V (PV charged) power. I can switch it off or on at any time- if the engine is running when the processor wakes up, it takes control but leaves it running. Likewise, if the remote control panel in my distant shop is switched off or loses power, the Lister controller assumes "carry on as previously directed and monitor".  In the last 5 years, I've never had a glitch or processor failure of any kind, so I really like the PIC/Picaxe chips.

After the experience I have now, I would now consider more fail safes, but the reliability of the controller and pneumatic controls has been so good that I'm not in a hurry, and other projects have priority.

[carbon-rod]

You are over thinking this thing...use energize to run..

Lloyd

The system would need to be directly connected to the spring loaded flap with the solenoid running continuously in order to open the flap. Basically it needs to be fail safe so that if the control circuit craps itself it will just shut down automatically and not require 12v to be injected by a control circuit, also if the power cable broke to the solenoid while it wasn't activated when you went to try and activate it for an emergency shutdown it wouldn't work.

On an energize to run system, everything has to be working else the system won't start. So it goes like this, the fuel rack solenoid only opens when in the run position, now the over temp, and low oil are also wired in this system, along with over/under speed, then so is the butterfly valve on the intake(you can get one of these from Granger) the butterfly valve can run on the same linkage to the run solenoid for the fuel rack.

So a fault at any system monitor ie over/under speed, low oil, high temp, de-energizes the run-solenoid, which then closes the fuel rack, and the intake butterfly valve, if you want belt and suspenders I would use a separate solenoid for the butterfly valve. I would also put a temp sensor on the exhaust.

Lloyd

That is almost exactly what I am planning.... I think it was mentioned on one of the previous pages.

The only difference is there will only be one solenoid to shut the intake flap / butterfly valve (good idea, I might use a butterfly instead of a flap) the rack will be driven by an RC servo so that will already be controlled electrically, if for some reason the servo is unable to control the rack ie rack is binding or servo fails then the secondary form of protection will shut the engine down.

Bruce that does sound like a great setup you have there, unfortunately for me I won't have an air supply to use for starting / actuators so I might end up having to use an electrical method instead to initiate starts. It is unfortunate because a linear actuator wouldn't come close to the same functionality of an air cylinder to push the wheel against the flywheel.

I will be using an Atmega32 processor most likely, plenty of IO for sensors etc..