Diesel instrumentation
 

I going to get a 1985 Non turbo :turtle:diesel Mercedes 190D 5spd.

What instrumentation is needed for monitoring best fuel economy?
Exhaust gas temp is a given but what others are needed. Would Vacuum gauge be a help?

Thank You

The only thing a vacuum gauge will tell you is if you have any restrictions in your intake tract that cause flow backup... and a flow meter will tell you better.

Diesels don't have throttle like gas engines do, so they don't generate vacuum unless there is some other form of restriction.

I see you're in MD - check york, lancaster, and harrisburg PA craigslists for a deal on 'Cedes right now... 3 of them for $1800 (I think they're all 190D's as well.)

I don't know if anyone has successfully adapted an mpguino to monitor a diesel, but I think consumption on those older ones is just a function of lever position and rpm.

Ummmm

For an NA diesel an EGT gauge isn't going to do much good. About the only thing that would do much good for a non turbo diesel is a barometer and an intake air temp (IAT) gauge.

Don't drive on hot days when the barometric pressure is low.

Edit: Make use of your rear view mirrors to to make sure your not blowing any black smoke.

I am in the same position as you, with a non-turbo, mechanical pump diesel. As far as I'm aware there is no way of monitoring 'on the fly' fuel consumption accurately.

You could invest in a diesel fuel flow transducer (commonly used in boats), but all this does is measure what is drawn from the tank to the injector pump, and then subtract what is returned to the tank - though these are expensive, and are not very accurate on low flow rates.

I have found what is commonly called the "power screw" on Bosch injector pumps, and progressively turned it down over a few tanks. I have less power for sure, but economy has increased, and smoke is reduced also.

If you are going to buy any gauges, look into coolant and oil temp - my engine was being overcooled and wasn't operating effeciently - I have grille blocks in place now and gained another few MPG.

Other than that, the less revs and throttle the better! Hope some of this is useful!

Been driving a 01 Dodge Cummings Diesel 6spd 2500 4x4 quad cab (almost 8000 pounds empty) with dynoed 383 hp and 916 ft# to the ground. I kept every drop of fuel for the first 118,000 mile and averaged 20.28 mpg. If you keep egt below 600 degrees it gets better fuel mileage. I'm in the middle of the best tank since I just found this site. Truck now has 174,000 miles on it and still regularly get 20+ tanks.

I'm looking forward to getting much better fuel mileage with the 2.2l diesel.

that 174K miles on a diesel is just scratching the surface when it comes to wear on a motor when it comes to diesel's.

i would like to also know how to accuratly measure MPG on a diesel. my bro has a 86 burban 2wd 6.2L. with a hummer intake mani, new injectors and some other stuff. he just goes from tank to tank to measure his.

2.2 diesel in 190D? That one wasn't availble in Europe.

Take a look at the specs:

http://www.mb.auto.pl/dane_en_os/229030DT.HTM
http://www.mb.auto.pl/dane_en_os/229030SK.HTM

Unfotunatelly no data about fuel consumption. FE data only for 2.0:

http://www.mb.auto.pl/dane_en_os/229010SK.HTM

And here is the ranking for 190D:

http://www.spritmonitor.de/de/uebers...e=80&gearing=1

Hope that's of any help :)

Sorry to say, but you can just forget about it. I looked long and hard for something for my 95 6.5L diesel and nothing exists. It isn't until OBDII arrived that it even became possible, and your 6.2L is well before that.

As Novaz nz said, you'd have to use mass flow sensors on the supply and return fuel lines and electronically subtract one from the other...all one-off of course.

I have suggested measuring throttle position and RPM to get a relative measure of consumption. Then combining those signals and a speed in an mpguino like device for display. Can anyone think of why that is worse than nothing?

For an older non-turbo, non-OBDII diesel, I think I would have to agree with you Gravedigger. Keeping your foot out of it and monitoring your EGT would be a way of helping with fuel economy. 600 degrees is a good start, but you'd have to experiment to find your 50 mph (+/-), flat-road EGT. Then keep your EGT at or below that, even on acceleration. My happy spot on my 96 F250 PSD just happens to be 600 degrees also. I'm not sure what else you can do with it for instrumentation other than an Inlet Air Temp (at the filter) or air pressure/vacuum on the inlet manifold (?).

Obviously, you know the tricks for HP. I'd at least change the intake setup and maybe the exhaust if it has a cat and/or muffler. I'd love to have a little diesel to play with!

The reason there are no mpguino-style gauges for older diesels is because the recirculating fuel system makes it impossible to measure how much fuel the engine is using at any point in time. Noting that there is a relationship between throttle and rpm doesn't change that. How would you calibrate all the different combinations of throttle and rpm to their fuel use - that's the original problem.

older (mid 80's) diesels dont use a TPS either. their throttle is flow control of the fuel output. and that regulates RPM.

From the diagrams I saw, it should be linear. The largest share of the return fuel is from the feed pump, but what gets forced through the injectors is controlled by the position of the control sleeve (controlled by the throttle) and the plunger moving back and fourth.

One could describe the control lever position to control sleeve with a function, or make a good guess at the demands at idle and at full throttle and come up with a formula that way. Alternately one could pop off an injector and measure how much fluid comes out of the injector after X revolutions for different throttle positions and map it that way. You might have to add a tps and an rpm signal though.

http://ecomodder.com/forum/showthrea...html#post63034

http://www.cs.rochester.edu/u/jag/vw...ARVESmall2.GIF

Ok, maybe not :) Those tension lever things might have some springiness in them and be active with the flyweights. So control sleeve position is a function of rpm AND throttle position (and "smoke screw" position to a lesser extent).

Hmm. If you could get a fix on the control sleeve position, it would be simpler. Otherwise you have to measure and build a rpm/tps to gph fuel consumption map.

Hmm... the piezo fuel line tach sensor approach is looking better.

You can derive rpm. You can probably also tell how long the pulse was, which is a function of rpm and control sleeve position (effectively stroke). Still need to determine the range but should be able to get a relative reading anyway.

fuel delivered = rpm * pulse duration * k

But how are you going to measure pulse length? I think you are confusing new, OBDII engines (for which gauges are no real problem and determining pulse length is relatively easy to do) with older engines. The older diesel engines like the 190D, for which there are no gauges, just use mechanical injection timing.

If you have electrically controlled injection timing there is already an Australian company that makes a device that measures the length of the injection event to determine instantaneous fuel usage data. I can't remember it's name but I found it on Google a year or so ago. I posted about it on a 6.5L diesel engine forum, so if anybody is interested I can probably find it in their archives..., but it won't work on older diesels with mechanical injector timing.

It is something DieselJohn had mentioned a while ago. The steel fuel lines actually expand during the injection on the old diesels, and it can be sensed with a clamp on sensor that came with some tachometers.

Edit: found a reference, don't know what the signal looks like though.
http://www.onosokki.co.jp/English/hp...ve/ge1200.html

Those steel-tubing fuel lines expand!? I suppose that's possible, but it sure wouldn't be by much.

Oh - I read the reference. The sensors measure vibration of the fuel line, not expansion. That seems more doable. You'd have a lot of work to do since it counts vibration events - not length of the events. Its response time appears to be 0.5 seconds, which is way slower than you would need. Still, you might be able to use some sort of piezo-electric sensor with different electronics. You could also use a microphone to detect the sound of the injector firing (which in its own way is detecting vibration). This is getting so complicated it might be easier to just buy a new car with OBDII.

"The sensors measure vibration of the fuel line"
Nope, not according to Diesel_John. I found his original post and he says he saw the whole pulse on a scope:

http://ecomodder.com/forum/showthrea...html#post34033

A lot easier to clamp a piezo on a line than to retrofit obdII if you happen to otherwise like your old diesel :). And even with obd you are not guaranteed fuel consumption feedback, I've heard a few reports about MPG on scangauges not working on diesels. So worth exploring a bit more anyway. Anyone want to donate an old vw diesel to me? :D

I don't know what Diesel_John thinks, but the information posted on the manufacturer's own site says the meter measures vibration:

For easy engine revolution measurement on 4-cycle diesel engines by clamping the piezo-electric type vibration detector to the fuel injection pipe.

I read the English challenged site too, thanks. I trust Johns interpretation for now, or at least that the pulse width is somehow ascertainable.

The 85 Mercedes was junk but I have purchased a 89 2.5l Auto. I wll be installing a egt gauge Thanks

Two different instruments. The piezo I was talking about was a timing light for diesels which I used to verify diesel timing before each sound test. Although the piezo lined steel ring was designed to trigger a timing light from an injector line, I looked at the raw signal on a scope and it clearly showed the entire injection pulse. True a piezo pulse does decay over time and true there is ringing on top of the signal caused by the needle bouncing on the seat in the injector among other system resonances, never the less the entire injection event can be discerned.

Cool, so if we know pulse width, and rpm, do we have enough info to figure out how much fuel was delivered with that pulse (at least relatively)?

I mean anyone can cut a strip out of a radioshack piezo and strap it to a pipe with a hose clamp :) (might need insulating in spots)

As i remember the steel pickup transducer was very stiff maybe 1/4" wide and thick. Found a photo. Looks like two semi circular chunks of piezo with a stiff collar backing it up. Loosening the thumb screw allows the piece to split. The contact area to the line needs to be paint, grease, dent, moisture etc. free. Available in most line sizes for about $60. or make your own. Charge voltages are high so a circuit to change to charge signal to a voltage signal would be good.

Actually vibration is not a good way to describe it. The transducer really does pick up very small pulses of the steel line expanding when the injection even occurs. You are talking about alot of pressure here and it happens pretty fast. If it was vibration then what vibration? because the whole engine has tons of various harmonics and vibrations.
With these older engines the easiest way is to use two zemco like flow meteres and subtract the return line pulse from the supply pulses, take the leftover pulses and feed those into the mpguino. DCB could, I am sure modify the program to count pulses and associate fuel use to each pulse similar to the way we do now. Shoot with my peak and hold gasser injectors thats basically all the mpguino is doing anyway.

Ugh, its a case of manual being written by someone in another country. The manufactures site is WRONG. Tell them to re-evaluate that statement and run it by their engineers to get it fixed.
Maybe it got off in a translation or something but that is not the way these transducers work.
If the meter measured vibration it wouldn't work at all for the application because of the fact that there are tons of little vibrations from just about everything. Translation could easily render the pulse produced by the pump as a vibration. If the tiny bit of tube expansion that is picked up by the transducer could be considered a vibration it is still not a good word to use to describe the event. Hitting a gong causes vibrations. A tuning fork makes vibrations. The pulse event from an injector is just that. A pulse event. Not a vibration.
I went to school for this stuff and worked as a diesel mechanic some 15 20 years ago.
I have my doubts that you could pull off a signal good enough to calculate fuel usage from one of those pulse events.

Interesting.. And did that signal change based on load? Thats the test right there. Is it possible for you to capture a idle signal and post it along with a full load signal and post that image as well?

The premise is that the old timey diesel pumps are positive displacement. If the length of the injection event can be determined as well as the rpm (both from the same sensor) then it should be possible to get a relative consumption figure of the fuel that was just squirted into the cylinder. That would be the simplest in terms of installation anyway, but you might have to go through a few tanks to get it dialed in (business as usual for most fuel consumption gizmos)

Zemco like flow meters... where to find them?
 

It would be great if you can share with all of us where we can get "Zemco like" flow meters.

Thank you in advance.

Oldbeaver.

hmmm what about fuel leakoff? This is literally fuel that leaks past injection pistons at the pump and at the injectors. This is the wear factor and will change with time. I guess you could recalibrate once a year or something. Also what about differences in pump design? I am only really familiar with the stanadyne pumps that came on early GM vehicles. Basically the piston rolls up a ramp. At idle it barely touches the ramp where at full throttle it is allowed to hit the whole ramp. Timing is controlled by turning the ring like housing that has the ramps in it. I could post a picture of the parts that actually do the injection events. I have a stanadyne pump from an GM 6.2 liter diesel. These pumps were used a lot and eventually modified to be computer controlled for the later 6.5 engines.

i was just toying with a prototype injection signal measuring sensor yesterday, you might need to attenuate the signal a bit with a couple Meg pot so you don't over range your A to D.
Parts: 2 used GM knock sensors, one 1/4" FPT x 1/4" FPT pipe coupling.
Tools: a drill bit the size of my injection line, a 1/4" pipe tap, hack saw or dremel.
Procedure: drill a hole in the center of the coupling, thread the coupling deeper on both ends until the knock sensors will reach in to the hole, cut one side of the coupling out until it will fit over the injection line.
Place coupling on a straight section of the injector line.
Hand tighten the knock sensors until they clamp on the injection. Wire the sensors in parallel. Attenuate signal as needed. see pic
i tapped on the bolt with a pencil and it made the analog bar on my Fluke jump, i don't think signal strength is going to be a problem.
Oh, BTW, there are two of these knock sensors threaded into the bottom of the water jacket on most injected 350 chevy's. make provisions to handle the antifreeze when you unscrew them.

Beav, i have seen them on ebay type sites. i try to make stuff. but if you don't mind buying, you can get modern flow meters for a $100 or less. An old Zemco will probably need a new ball and light, so can't go to high on price.
A lot of talk about it in this thread
http://ecomodder.com/forum/showthrea...-new-post.html

Dang John, you make it look easy :)

I'm not %100 on why two, or how to wire them up though. In series I presume, for a larger signal?

Oh, and obligatory drawing of what came to mind when I saw the pic :) Harder to manufacture but should be able to clamp down harder w/it.


http://ecomodder.com/forum/attachmen...1&d=1240357794

P.S. we should probably keep the diesel stuff in this thread. Hate to see you guys having to type everything in twice.

two opposed because they cancel horizontal "vibration" but retain the expansive signals. wire parallel for this effect. i noted there were two in the store bought example. GM because there is an endless supply. don't squash the line, hand tight. difficult to wire in series anyway, base is one wire and connector is a single pin. tools to harvest sensors 7/8" 6 pt. deep socket (they are very tight), wire cutter and a bucket.

Ok this is just ingenius. I love it.
questions and suggestions. Id love to see a scope reading off of this.. part and full throttle.. pretty please?
I suggest a junkyard for those knock sensors or just get a couple new ones from the parts house.
I seriously would drop all my current projects to try this except its a pain in the butt getting to my injector lines in my VAN. I have a custom doghouse and laptop holder all setup right in the way.

Im sorry but you cant get them anymore. They are out of business. You can sometimes find old ones on ebay but not very often. I have fixed a few old ones so even if they dont work and you find one keep it. They are very simple. Most dont work well on larger engines without slight modifications, due to the small orifice inside.
Of course you still need a circuit board design to drive it and produce a signal the mpguino can read.
DCB or one of the others may be able to post a schematic that would be able to render a signal suitable for the mpguino.

If you did it this way dcb you should probably back up the injector line or it might get bent. perhaps weld in a flat piece of steel to back it up or better yet weld in something the shape of the injector line for full support on the back side. Then thread one sensor on over the top. What do you think diesel John?

Find a flowmeter first, then we can discuss :)