Modeling a restrictor in WAVE

[Wildcat13]

Here at the University of Arizona we are starting to to some WAVE modeling. After going through the tutorials I tried my hand at modeling a restrictor. I don't know if it is due to limitations of the program, or limitations of my skills, but I have been unable to get the restrictor to choke. I was successful in getting a Ma=1.4 at the throat of the nozzle by changing the pressure ratios and divergence angle, if you can call it successful.

Any suggestions from those Ricardo buffs out there about common mistakes that might cause this?

Ryan Kanto
University of Arizona Formula SAE

[Ben Inkster]

What engine speed are you using? Try different speeds and compare the change.

[John Lankes]

A couple of things to try:

1. Did you actually model a restrictor section? If the section length is less than the discretizatoin length, this can cause a problem. We model the restrictor as 2 duct section, one converging to 20mm and one diverging from 20mm.

2. Do a sweep plot of your air consumption (airkgh) versus RPM, and verify that it is in the believable range.

3. How did you model your actuall cylinder head ports? Modeling them as an inlet and valve exit diameters generally works best. Modeling them as a series of duct sections can cause odd problems.

[Wildcat13]

This was done as a standalone test without an engine. A pressure drop was put across the restrictor in one set of cases (sweeping from atmospheric to something like 0.6 bar). The restrictor was modeled as 2 ducts connected by an orfice, with the discretization length being smaller than the duct lengths. Flow in one of the cases is over 400kg/hr.

Ryan

[Charlie]

WAVE is not supposed to be accurate for angles > 5 degrees. At least the converging angle on your restrictor should be above this, and probably diverging too.

If you try to use WAVE to optimise your restrictor design you will run into this issue and get very flawed results.

Good job in trying out a simple WAVE model of just the restrictor instead of doing an entire model and wondering why it's crap! However this is the limitation of the 1D model.

There are several ways to work this; my solution was to measure the presure drop across the restrictor on the running engine (dyno pull), and have a variable orifice in WAVE that matched the pressure differential at all the RPM points. Seemed to work OK.

[DaveC]

I wrote an email to Ricardo's tech support on this, and was told "10-20 deg. should be OK." I just modeled the restrictor close in form to the actual venturi, and lessened the taper angles (increased the duct legnth) a little to run < 10 deg. taper. It seems to work OK, but power does not fall off after the power peak as sharply as it actually does. Charlie, do you use a smaller orfice than actual as engine speed increases? Has anyone used Vectis for the restrictor to come up with more accurate results?

Also, I am looking for a way to change intake camshaft duration and lift (not overall timing, I know that is done with the "intake anchor") using the constants panel, as in the tutorial experiment. Its not clear to me how to do this without making up new valve lift curves in Excel, and testing the curves individually.

[nathan s]

From my experience, Ricardo prefers to have the restrictor 1 sided, as in 20mm on the front, and your exit diameter all the way across on the back side. At least, that is what Ricardo told me.

As for engine speed, I think, from some hand calculations that we did, choke flow should be between 10 and 12k rpm.

[DaveC]

From my experience, Ricardo prefers to have the restrictor 1 sided, as in 20mm on the front, and your exit diameter all the way across on the back side. At least, that is what Ricardo told me.


Is that modeling the restrictor with a complex or simple y-junction? I have it modeled with two tapered ducts connected with an orfice right now...

[Charlie]

I could say exactly what I did but it won't neccessarily do you any good. Different engines and intakes, etc. If you want a good match then take data from a running engine and match your model. Look at the pressure drop across the restrictor. Fudge it until it matches what you see on your running engine.

If you don't have a running engine to test on, then you are wasting your time with a simulation model.

[Mustang Mac]

Is that modeling the restrictor with a complex or simple y-junction? I have it modeled with two tapered ducts connected with an orfice right now...

From my experience, you shouldn't model anything as a simple y junction. Always use a complex.

[Wildcat13]

Originally posted by DaveC:
Also, I am looking for a way to change intake camshaft duration and lift (not overall timing, I know that is done with the "intake anchor") using the constants panel, as in the tutorial experiment. Its not clear to me how to do this without making up new valve lift curves in Excel, and testing the curves individually.

Dave,

There are multipliers for the duration and the lift under the valve profile settings. All you have to do is change these to constants and vary accordingly.

Ryan Kanto

[jon109]

hey

i am using ricardo wave for my thesis project. In a few words i am simulting an FSAE kawasaki engine so as to finally try to tune the valve timing etc... i am having problems to model the restrictor.

I've already done experiments to get the characteristics of the restrictor (i.e mass flow versus downstream pressure). but now i would like to use the data i have in my model. I also found the respective orifce area to allow the same flow at resopective downstream and upstream pressures. so here i derived a 2D interpolation map of orifcie areas for the respective upstream and downstream pressure. but unfortunately when i tried to implement it on wave, it didnt work.
can anyone help me to find a way to model the restrictor well.

note: i also tried to model it in wave by a nozzle and diffuser and did a flowbench test on it. apart of taking long times to solve, it is not as accurate since the diffuser modelling is quite hard to get in wave. i also tried to increase some pressure loss and friction coefficients so as to get the right flow characteristicts. Hopefully there should be something more professional to do this...

tnx in advance for any help http://fsae.com/groupee_common/emoticons/icon_smile.gif

[Kamil]

It seems strange that a variable orifice diameter didn't work for you. Did it give you an error or rubbish results? One way of modelling i've tried is to represent the model as a series of constant area, stepped ducts, as opposed to a tapered duct. By changing the length of each section and size of diameter step, you can tune your restrictor model to give you the correct plenum pressure as on the dyno.

[John Lankes]

Is there a good way to look at average pressure directly in Ricardo? I have been converting the cyclic pulses to an average pressure, or putting the values into excel and taking an average for each rpm case. I was wondering if anyone had a better method, or perhaps I am just completely missing one of the plot parameters.

[jon109]

yes you can fnd the average pressure for a given speed by plotting a sweep plot of duct pressure (dpl or dpr) versus speed.

[jonny]

Hi, I modelled a restrictor using an orifice with one duct diverging to 20mm and the other duct converging from 20mm.... I'm not sure whether this is the problem but i get negative values for my torque and my bhp!!!! I have used some data from the computer ie Valve FLow coefficeint, Burn duration and such...

Anyone has same problems as me??

[Krautsalat]

If you get negative values for work and power there is usually something wrong regarding the combustion. During the simulation the engine is run at a prescribed rpm and without combustion there is only the negative work from friction and pumping losses. So you should check if the combustion takes place(mass fraction burned gas) and if the timing is right(start of combustion). Did you look at the cylinder pressure during a cycle? Another possibility would be completly wrong valve timings or so. Is enough air entering the cylinder during the scavenging?

Hans

[samphlett]

If you have only changed your restrictor (not combustion, fuelling or valves) then it's unlikely to be a combustion problem - more likely a breathing problem. Check the cycle-averaged pressure throught the path from intake, through the restriction to your intake manifold.

[jonny]

I have managed to solve the problem of the negative work done, its due to an excessively long intake runner which in the end caused the engine not to take in any air...
Now there is a more peculiar problem, from 6000rpm to 8000rpm there is a sharp decrease in my torque, from 24Nm to only 4Nm!!!! the graph then goes up back to 10 Nm at 10000rpm... I'm clueless...

[samphlett]

Bear in mind that the FSAE grant includes some support hours. Why not talk to your faculty supervisor if having problems with your model? He/she will be able to route questions to the WAVE support team.

[jon109]

note that teh dip in performance of these restricted engines is common! the problem is the incorrect valve timing which is originally designed for a high flow engine! if you can manage to get dyno data of your engine you should see this dip aswell!

[jonny]

just to share further, I am getting some low torque at low rpm about 24-30 Nm from 4k rpm to 6k rpm... then there is a sudden increase of torque from 30Nm to 50+Nm at 6k rpm to 7k rpm.. This torque stays more or less constant but dips at abt 10500rpm..
I modelled the restrictor as t pipes joined by an orifice, the orifice of diameter 20mm... My intake lenght is about 350mm, pri_ex~ 650, sec_ex~ 770... Have been trying many iterations but the low torque for that rpm range is still there...
Anyone with any ideas on how to flatten up the torque curve a little are welcomed to give their vies...
Thanks in advance

[Chris Allbee]

Looking at your intake length (providing thats the entire effective length including head) it would have a peak at ~9500 to ~10000RPM, this would give a trough in the torque shortly after the peak...

[mech5107]

Finally, which is the best way to model a restrictor in Wave?

I'm using a 20mm orifice with tapered duct to the plenum...