Odd Sensor Design

We have an engine model which is prone to a particular problem. When the engine goes down after it fails a turbo charger, it slowly fills the aftercooler and intake manifold up with oil. When the engine shutsdown, the aftercooler drains the oil into the manifold. When a operator tries to restart the engine, the cylinders fill up with oil (only after it actually starts) then hydro locks and spits a connecting rod out the side of the block. {yes, the engine generally starts fine with no hydrolock until it starts coming up to speed, then it sucks the oil in and BAM!}

This is an industrial engine, which runs un-manned 24/7 and has a shutdown/safety panel. We have several hundred of them.

What we're looking for, is some sort of sensor that can detect presence of oil in the intake manifold. Oil can reach up to a 1/4" deep. We have a port we can screw into, now we have to determine the best way of detection.

Since oil is a pretty good insulator, it will have an affect on what type of sensor to use or design. Items we thought of....

Float style or tip level sensor - Hard to read a 1/4" level Small, low temp hot wire, measure the current, if submerged in oil it may change, vrs air/gas mixture Capacitance measurement? - Possibly the best method???

The method can not cause combustion of a combustable mixture. We have 24 vdc on skid. Normal shutdown method is to pull down inputs from our panel (3.3 vdc TTL with pull ups) grounding an input forces it low thus a fault.

Any clever ideas for this one?

Richard

----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----

formatting link
The #1 Newsgroup Service in the World! >100,000 Newsgroups

---= East/West-Coast Server Farms - Total Privacy via Encryption =---

Reply to
Richard
Loading thread data ...

I'll look at the link you sent in more detail. Thanks, RIchard

----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----

formatting link
The #1 Newsgroup Service in the World! >100,000 Newsgroups

---= East/West-Coast Server Farms - Total Privacy via Encryption =---

Reply to
Richard

On this particular engine it's not possible. The aftercooler actually sits in the intake so there is no way to catch the oil "before" it drains into the manifold. Also, the manifold is integrated into the engine block "Vee" and there are no taps anywhere low enough that would allow that.

That actually was our first idea, until we looked closer.

Thanks though, Richard

----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----

formatting link
The #1 Newsgroup Service in the World! >100,000 Newsgroups

---= East/West-Coast Server Farms - Total Privacy via Encryption =---

Reply to
Richard

the

shutsdown, the

to

actually

the

until it

presence of

have a

detection.

type of

may

vdc TTL

Thermistors are good for this sort of work - read their resistance when they are dissipating 10uW (0.01C of self-heating in air) and again at

1mW (1.0C of self-heating in air) and if you can't detect a 4% lower resistance at the higher dissipation, the thermistor is immersed in oil or water ...

Betatherm make mounted thermistors specifically for this sort of job

formatting link
(Immersion)

-------- Bill Sloman, Nijmegen

Reply to
bill.sloman

Nasty problem. Google for capacitive oil sensors.

Regards,

Boris Mohar

Got Knock? - see: Viatrack Printed Circuit Designs (among other things)

formatting link

Reply to
Boris Mohar

The thermistor idea is a good one, but probably requires a controller, due to the wide temperature (and hence resistance) range of an industrial engine manifold. You could instead do something with two (mildly) hot nichrome wires, one near the bottom of the manifold and one half an inch above it. Wire them in series, and measure the centre tap in a bridge arrangement--if there's a significant imbalance, you have an oil problem. That method needs only an amplifier and comparator, and will work over a 200C temperature swing without too much difficulty. Thermocouples would work, but the signal will be pretty small, leading to potential noise problems.

Cheers,

Phil Hobbs

Reply to
Phil Hobbs

What (approximately) is the temperature and pressure range at the sensor? I think this could be done very reliably at a reasonable cost using the thermal conductivity principle and Pt RTDs.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
 Click to see the full signature
Reply to
Spehro Pefhany

the

Hmm. Remember the intake manifold is under a variable vacuum - getting oil to flow into a resevoir (especially from an area of high turbulence) may not be as simple as it sounds.

Also the OP has stated there is no room for fittings and hoses.

Reply to
Mark Jones

Thread a little tube over to where the oil will most likely converge and secure it somehow (magnets, epoxy, etc.) Then run this tube to a small pump, like:

formatting link

Pulse this pump every few minutes or so. A sensor in the "output" vessel can then easily tell if any oil has accumulated and if so, latch the pump on continuous and trigger shutdown - both removing the oil and shutting down the engine.

p.s. I'm looking for work. :)

-- "If necessity is the mother of invention, then is experimentation the father?" MCJ 20041116

Reply to
Mark Jones

Would a 'squing' sensor work?

Reply to
s--p--o--n--i--x

This would be a good starting point:

formatting link

sPoNIX

Reply to
s--p--o--n--i--x

I read in sci.electronics.design that Richard wrote (in ) about 'Odd Sensor Design', on Tue, 18 Jan 2005:

Could you connect an oil-drain reservoir to this port, so as to drain the oil away and eliminate the problem at source?

--
Regards, John Woodgate, OOO - Own Opinions Only. 
The good news is that nothing is compulsory.
 Click to see the full signature
Reply to
John Woodgate

If the port is below the expected oil level and it is possible to attach a glass tube or transparent plastic tube to the port, then you can either have the operator check for oil before starting engine and/or use a simple optical sensor to disable engine start.

__________ / \ | intake | | manifold | | | \________ _/ | | [| |]

Reply to
Goran Larsson

the

small

Fitting everything through a 0.25" orifice is a bit constraining (kind of like being a feline gynecologist, I suppose). But that's what makes it fun.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
 Click to see the full signature
Reply to
Spehro Pefhany

On Tue, 18 Jan 2005 06:05:46 -0600, "Richard" wroth:

I would find the lowest point in the intake system where the oil can get to and make a hole. Then I'd make a hole at a point in the intake way above the highest level the oil can reach. Then I'd put fittings in each hole with small tubes leading down to a sealed container just large enough to hold most of the oil that leaks out.

That way the engine will be protected, it will always be able to start and run, and the added container can have a normal float switch because, if it is the right size and shape (taller than it is wide), the level change there will be big enough. The float switch doesn't have to prevent starting, just light a "check engine" light.

Jim

P.S. Two holes and two tubes are necessary so that the oil will flow easily down the lower one while the air in the container can go back to the intake through the upper one. No "oil lock".

P.P.S. Sometimes I'm so smart I amaze myself!

Reply to
James Meyer

detection.

How about a thin pibe with one open end close to the bottom of the manifold and the other connected to a container where the oil will accumulate, then measure the level in the container using a standard automotive oil-level level sensor?

The container could be shaped to give a good reading for little oil, i.e. tall and narrow. Normally, a little air will escape through the pibe, when the turbo begins to fail oil will be pressed through the pibe too, accumulating in the container and triggering the alarm.

I assume that normally, there will be over-pressure in the manifold when the turbo is running and that - when the turbo fails - oil will leak out at an increasing rate; initially being burned by the engine but eventually the bearings of the turbo will seize and that's where the engine dies.

Reply to
Frithiof Andreas Jensen

I developed a thermister based level detector for a long extict project, if you could get a suitable thermistor it could be finished to suit your purpose. If you cant find an "off the shelf" device email me your address and i'll send you

1 for evaluation.
Reply to
CBarn24050

Wow, lots of good ideas. Manifold pressure can be as high as 20 psig with the temperature ranging from 130-180 Deg F while running. We actually try to hold it at 130F, but depending on load and ambient it can vary.

Thanks, Richard

----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----

formatting link
The #1 Newsgroup Service in the World! >100,000 Newsgroups

---= East/West-Coast Server Farms - Total Privacy via Encryption =---

Reply to
Richard

I suppose that is the best method, however Caterpillar has worked on it for about 4 years now. All Turbos eventually fail, so that is a given, and redesigning the engine block and manifold, well, I doubt we will see that ever happen. Problem is, we have very many of these model engines!

----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----

formatting link
The #1 Newsgroup Service in the World! >100,000 Newsgroups

---= East/West-Coast Server Farms - Total Privacy via Encryption =---

Reply to
Richard

Oh, that is the easy part. Again, a turbo charger is prone to extreme temps and speeds, they all wear out. Yes, prevenative maintenance can eliminate much of this by taking axial and radial clearance measurements on a routine basis and change "before" a failure, but not failures give you such an advance warning.

Kind of like having a blow out on your vehicle. Sometimes you can prevent it by keeping an good eye on your tires and properly maintaining them, other times, it just happens. When it does, well, hopefully you can pull over and change the tire, or to some people, they loose control and drive off a cliff.

I don't consider a safety shutdown to save a $100,000 engine as a bandaid, besides, there are approx. 40 total safety shutdowns for one of these engines. Surely they aren't all considered bandaids. They are there to shut the engine down in case something is abnormal or something breaks. Since they are industrial engines and run un-manned they are needed.

I guess we could take all the safeties off, and then have Cat build a perfect engine. Heck, then we wouldn't need technicians .

RS

----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----

formatting link
The #1 Newsgroup Service in the World! >100,000 Newsgroups

---= East/West-Coast Server Farms - Total Privacy via Encryption =---

Reply to
Richard

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.