Completion of engine test bed Direct Reaction test!

[Flytox]

Hello Camel1

You can go into details, like putting photos, the diagram of your hydraulic brake, how exactly does your load control etc ... that we can start to cogitate

[camel1]

Hi everybody !

...
You can go into details, like putting photos, the diagram of your hydraulic brake, how exactly does your load control etc ... that we can start to cogitate

Okay, I will already provide a crobard which explains the current installation (manual) and a draft of an automated version:



In the current version, the valve at the pump outlet makes it possible to set the restriction on high pressure, and read the result on the pressure gauge.

In the automated version, the idea, coming to discuss with Christophe, would be to set a setpoint with a motorized proportional valve, then control variations thanks to a series of all or nothing solenoid valves, each having a nozzle of different diameter.
For example, with eight EVs, having progressive section nozzles, in power of two, one could obtain 256 different leakage values.
The advantage of EVs is a much quicker response time than a proportional valve ...

Finally ... it's an idea, I'm not really competent in hydraulics, that's why:

...

A hydraulic circuit controlled by a PLC should be feasible in your case! But again it's expensive ... in money and time!

Varying a hydraulic load is quite simple, especially if you have already designed the engine brake system using this energy.
For example, with a solenoid valve and a pulse generator, you can vary a load and it all costs around € 200 maxi ready (we have that at work).
Cdlt

Apparently, you know something about hydraulics, pb2488 (funny nickname ... )

Could you give us more info on your solution?

To continue responding to Flytox, we adjust the load, as I said, manually.
That is, a lever on the pump valve, and a lever / dial connected to the injection pump to adjust the RPMs - knowing that when you touch the engine brake, you vary the RPMs ...

This means that if you slave the engine brake side, you must do the same on the engine speed control, so install a servo motor instead of the RPM lever, and connect it to the control interface .

When André makes his road tests, he stabilizes at cruising speed, and regulates "at the foot" to compensate for false flats, climbs and descents (load variation)

When Dede plows his field with his tractor, he shifts into third low gear, engages the plow, and pushes the accelerator lever to the bottom, it's the pump that regulates (I think?)

On the bench, an automation should be able to simulate these working conditions without problems, in a perfectly reproducible way ...

In short, there are some hardware elements missing, and add a control loop in my supervision software, which can manage both acquisitions, but also order anything, just ask it ...

As for photos, screenshots and others, I'm taking care of it this week, and would add that to my site, to fuel the debate

A ++

Michel

[Alain G]

Good evening Camel1

Why not simply use an electronic pressure gauge coupled to the computer and a manual graduated valve?

[Flytox]

Hello Camel1

There is another mechanical solution, for the variation of the load (to see if it is simpler more efficient and more economical than your solenoid valve solution).

At the end of your load control lever, you put a bearing which runs on a cam whose profile you invent to simulate plowing conditions etc ...

Your cam is driven by a continuous or stepping motor / reducer whose rotation speed you control perfectly (speed control). This can be done over a portion of a revolution (forward / reverse or by turning continuously and reproducing a cycle at each revolution. If the control is rigid, lubricated, this can be reproducible.

Another idea:
To imitate the "cubic characteristic" function of some power bench which makes it possible to regulate the engine load according to the engine speed cube (image of what happens with a car moving on the road) it would be necessary to find couple the engine RPMs to the load cam which would follow a cubic characteristic profile. Thus, there is no longer any risk of stalling the engine or of going over-revving due to improper handling / desynchronization of the load. There remains only the engine gas control to control load and RPM as on the car.

In any case, there is a drawback ..... this bench does not simulate the very large inertia of the vehicle in any cycle. It seems to me a parameter very important and particularly in water doping where the charge moves a lot. If you have 5 minutes ... you can't make one ...
But, no it's not even going to cut your budget and mortgage your vacation

[the middle]

If this load problem is bothering you so much, I have to move my ass, and see the old total research center where I was, on what principle they worked (there were a dozen benches!)
I still have some old friends
There was even a cfr engine (being able to play with compression, piston stroke, etc ... )
(I'm going to move my ass ) 15 days .. this vat?

[AXEAU]

Hello everyone.I have been interested in fuel savings for some time and I have just carried out a G.Pantonne installation on my 1.5d ax thanks to your work. Seeing the photos about this test bench, i 'would like to share with you some ideas, intuitions.

To start with, make sure that this engine is at its best by checking the calibration of the injectors, the adjustment of the valve clearance, the tension of the timing belt, the traces of scale in the adm and ech pipes of the cylinder head and compression measurement.

For the collection of information it is interesting to measure the mass of air absorbed by the engine with a flow meter and t ° + a second to determine the quantity which passes to the reactor.

Using the speed sensor and if the crown is notched all around, the efficiency of each cylinder is measured with an oscilo. Each explosion causes an acceleration of the rotation, a comparison can be made with / s steam.

Finally, it can be the hardest. I fear that the distance between the outlet of the reactor and the cylinder is too great and that the vapor loses a large part of its characteristics. leaves to shorten it, double it and add a spacer in front of the turbo to have the necessary length. The double outlet of the reactor would be done through the ech pipe. to arrive on a common ramp and then join the collector adm as close as possible the cylinder head by 4 conduits. The air of the reactor would be taken after the turbo with a device to vary the quantity. Difficult to imagine shorter.

It may still be to insulate around the temperature probes or the assembly and do the tests.

here are some ideas, if it can help you ... i have others and many questions too. So see you soon.

jlg

[Flytox]

Hello and welcome to the AXEAU club

for information gathering it is interesting to measure the mass of air absorbed by the engine with a flow meter and t ° + a second to determine the amount that passes to the reactor.

Do you have any idea how to make this instrumentation without leaving your shirt there?

Using the speed sensor and if the crown is notched all around, the efficiency of each cylinder is measured with an oscilo. Each explosion causes an acceleration of the rotation, a comparison can be made with / s steam.

The oscilloscope is one thing, after you have to interpret the signals, the best is a computer recorder. If Camel1 does not have one, it would usefully complement its instrumentation ....

It would be necessary to keep the gv downstream of the turbo and install the reactor upstream even if it shortens it, double it and add a spacer in front of the turbo to have the necessary length. The double outlet of the reactor would be done through the ech.pour pipe arrive on a common ramp and then join the collector adm as close as possible to the cylinder head by 4 conduits.

Move / reverse disturb the gas upstream of the turbo chui not sure it is a good idea. The average temperature in the exhaust collapses very very quickly with the distance of the cylinder head. The performance of the turbo will be very affected. I guess you have to keep the reactor behind, that it's a better compromise. André what do you think?

A+

[camel1]

Hi everybody !

I'm bringing back new photos and additional details this afternoon ... as promised!

...
For the collection of information it is interesting to measure the mass of air absorbed by the engine with a flow meter and t ° + a second to determine the quantity which passes to the reactor.

Welcome, good idea, so we would need two air flowmeters, it is in the breakage, must see the price ...
Electronically, this assumes two counting inputs, I no longer have any available on my LabJack, so, hack to add ...

... Using the speed sensor and if the crown is notched all around, we measure with an oscilo the efficiency of each cylinder. Each explosion causes an acceleration of the rotation, a comparison can be steamed.

Yes, I know this manipulation, we have already done it with the guys from the IUT, a "tooth sensor" produces a signal which provides information on micro - acceleration decelerations, therefore on combustion cycles ...
In fact this will be done in a second step, when we have stabilized a version of the proto which gives us a tangible and reproducible result, we will bring them back with their acquisition station, and they will put accelerometers on the engine and an encoder on the axis of the crankshaft, to be able to make acquisitions with / without steam.
They have the mathematical tools to extract useful information from this raw data ...

... Finally can be the hardest. I fear that the distance between the outlet of the reactor and the cylinder is too great and that the vapor loses a large part of its characteristics. It would be necessary to keep the gv downstream of the turbo and to install the upstream reactor leaves to shorten it, double it and add a spacer in front of the turbo to have the necessary length. The double outlet of the reactor would be done through the ech pipe. to arrive on a common ramp and then join the manifold adm at closer to the cylinder head by 4 ducts. The reactor air would be taken after the turbo with a device to vary the quantity. Difficult to imagine shorter.
It may be to insulate around the temperature probes or the whole and do the tests ....

Imagine that we have already thought about this thorny question, and that in general, the pantonist must always make a compromise for the placement of the reactor ... Far from the intake, and close to the exhaust, or else the opposite ? What should we favor?

Our previous experiments were on an athmos diesel engine, and here we are on a turbo ... the game is changed.
This is why we have provided different points of tapping, and different ways of connecting the reactor.
And to tell you the truth, we've already tried all of the options, with no tangible results yet.
Anyway, the advantage of the bench is accessibility, and the possibility of changing the design at will, without the usual constraints of the prototype on a car ...

Here for now, I would probably come back tomorrow, with a little update of the site ...

Thank you for your participation !

A + + +

Michel

[Capt_Maloche]

-An exhaust gas analyzer ... (Maloche? Cheesy Grin)

I have to equip myself, I had borrowed a friend's equipment

but it is in my forecast

[Other]

Hello

Move / reverse disturb the gas upstream of the turbo chui not sure it is a good idea. The average temperature in the exhaust collapses very very quickly with the distance of the cylinder head. The performance of the turbo will be very affected. I guess you have to keep the reactor behind, that it's a better compromise. André what do you think?

If we are looking to operate water doping at low load city driving, ideally this would place the reactor before the turbo, but in practice it is almost not possible with the design of current automobiles, at low power the turbo is practically inactive, but the exhaust temperature drops quickly after the turbo
so for reasons of simplicity we place it after the turbo

This is why we have provided different points of tapping, and different ways of connecting the reactor.
And to tell you the truth, we've already tried all of the options, with no tangible results yet.
Anyway, the advantage of the bench is accessibility, and the possibility of changing the design at will, without the usual constraints of the prototype on a car ...

for a long time I believed that the external temperature of the reactor was a main parameter but this is not the case (a reactor on an autos petrol engine is almost double the heat even at low speed however its performance is less good than on a diesel which only has 300c at the exhaust

Performance of doping with water does not increase with the temperature of the reactor or with the outlet reactor temperature

There are several way of seeing things
Some are a preheated before entering the reactor (Vitry) else do a warming after the reactor
else no preheating (HYPNOW)
I did a little warm before entering the reactor and reactor outlet heater (rather a detente output rod remaining in the exhaust.)

After several tests on water spray reactor entry
I pulverize me that idea about a hot water, even very heated regulated electric was not advantageous preferable to spray water at room temperature is a 25 30 a C and not to heat up this portion of bedroom, entry 22C low temperature has 30C gives hottest reactor exits as an entry 80c 90c.
My best performance also with reactor output 120C a140c
GV produces steam but adding air with this reactor inlet steam produces a condensation cloud fog
this fog must be in a lukewarm temperature neither too hot nor too cold if we want fine droplets.

I'm not Strongly believes that we must absolutely circulate it at high speed in the reactor, I eliminates the Venturi admission to less depression in the reactor therefore less speed, the water is still the consomation even the reactor outlet temperature to drop 180c has 140c
left to do some verrifications on fuel consumption measurements on the test runs.
Throttling in intake at equal power results in greater circulation in the reactor and higher reactor outlet temperature and higher exhaust temperature.
an (adjusted) reactor in an exhaust results in an exhaust pressure drop and an increase in exhaust and reactor temperature.
All these pressure losses intake and exhaust promotes the operation of the reactor at low power, but harmful on full power, so there is a reasonable compromise to make in our assemblies. (in my case I never need the full power and the speed limit on the roads is 100kmh, all the manipulations are done to seek the best output between 80 and 120kmh .. (the vehicle still ruins easily at 160kmh and easily maintains a 100kmh rating)
In your test do not forget to measure the exact temperature which enters the reactor and what emerges some cm before entering the collector (although measured with an internal sensor in the pipe to not suffer the copper tube contact effect that cooled in admission.
Another important measure is the exhaust temperature
it has linked the power demanded of the engine if no flange admission .. (she tells us instantaneously on engine performance)

Andre