Inlet trumpet dyno testing

[RUNAMUCK]

I like to regurgitate things that I've heard properly clever people say to make people think I'm a lot smarter than I really am. (But threads like this reinforce my suspision that I'm not that smart after all) My two cents on this whole thing, (Prepare a grain of salt?to dismiss completely)
A book I read once said that those full bell shaped trumpets outflow the old school style ones. (maybe inside an airbox, I dont recall exactly) 
and with the glorious video with the huge long PVC tubes, (Her indoors indignantly asked me to close the door if I'm going to watch loud videos like that)  I would wonder about the suitibiltiy of that diameter tube over that length? A 1600 is a pretty small engine. And the ID of that tube wouldnt be much (if any) smaller than the intake runners on a crysler 413/426 cross ram. (Thats knocking on 7000cc, or close to four and a half 4age's) With the trumpets at their shortest, it was just about getting the air in. But I would ask how fast the column or air was moving in the longest runner test/what was the velocity like? Ive seen a number of examples on the exhaust size where smaller diameter longer primary pipes make more power. Yes I barely grasp the concept of fluid dynamics. But I know that a 5 inch system on a 3K wont do much good. 

Not throwing stones, just agreeing with thats formerly been said that there are a LOT of variables to be considered. Also car makers have some very very clued up cunts working for them. and they have thousands of hours of R&D and testing under their colective belts. 

[Roman]

Velocity isnt fixed
Like, velocity at 4000rpm will be half of velocity at 8000rpm....
And dont be too generous towards OEM. On one of the Barra motors, it had variable valve timing, but the table in the ECU was blank. 
They just did all of their abstract validation bullshit and forgot about it!

And I remember seeing some pictures from when they were testing exhaust manifolds design for the FD RX7. 
they were just throwing shit at a wall and seeing what sticks, with some fairly oddball designs.

OEM has an extra layer of unhelpful complexity in that they're not trying to build one of something, they need to figure out how to make thousands apon thousands, in a way that uses the least amount of time and money.

So yes they have good R&D budget but their goals are very different too.

In my mind it's hard to beat real life testing. You dont need to know why something works, just that it does.

[kpr]

2 hours ago, Roman said:

 You dont need to know why something works, just that it does.

Along with.   " I don't know how it works, but I know how to make it work."    you have a winning combo     

[kpr]

 

new contender; the internal bellmouthed peach slices

[kpr]

done most of the testing over the last 2 days.   going to take awhile to put all the info together.   found some interesting stuff.

the peaches didn't do as bad as you would think haha

[kpr]

Here we go,   2 days of dyno testing and  even longer editing.    @Roman   also made some science graphs  with the dyno data.   maybe he will post them up here.

 

 

 

 

[Roman]

So for starters here's all of the results overlaid, the colour scale is length of the trumpet from 0-205mm




Then here's the same, but zoomed in over the important part which is 5000-9500rpm



Then here's the same as above, but only the trumpets shorter than 100mm



Then the trumpets over 100mm





Then, I looked through the raw data and for each run, did a "highlight results" in excel for the top 10 torque values. Then I averaged the RPM at which this happens. 
You can see that there doesnt look to be any sort of trend on the horsepower, but torque is a more relevant measure as the length of pipe helps maximize the cylinder fill. So wherever that happens you will have the most torque. 

So despite my blatherings about how I feel that harmonic stuff feels irrelevant when you see a lot of results, perhaps it isnt when you see the peaks and dips of peak torque rpm actually moving around back and forth like this. 



But it's worth noting that this is only a measure of where peak torque was made, not that it was any good at making peak torque there.

But maybe someone with engineering smarts like @ajg193 could draw some conclusions from it? I'm not sure what to think but does seem like a pattern. 

Havent tried to do any calculations for how this relates back to speed of sound or whatever, as of yet.

[kyteler]

Loved it.  

Awesome that the filter is best (unless I missed something). 

[kpr]

That intake with or without filter makes same power.  so yeh just ran it as is.   
 

Looks like if got all the lengths exactly the same. (since the opening shape and diameter have an effect)    the roman dave,  pvc with full bell  and my normal intake   would all be pretty much the same power throughout.       think the only reason my normal intake looks sightly better is its acting slightly shorter than the other 2   (dave +45 and the pvc w/ plastic bell)

 

 

[Yowzer]

Howabout a graph of best trumpet length vs rpm?

[Roman]

How do you mean? 

EDIT: Do you mean how much peak power each made, with each length? 

[Yowzer]

Trumpet length that makes most power / torque at any given RPM

[Roman]

Can you not see what you're wondering from the coloured graphs I posted above?
But, Peak kw doesnt change by a huge amount from that 65 to 185mm range. 
But the shape of the curve either side of it and the rpm at which it happens has a significant change.

Labelled my axes even, feeling fancy tonight

[Yowzer]

Yup I can see it in the above graphs, but I think it'd be interesting to see it mapped against RPM directly, ie how you'd map the control of variable length runners

[kpr]

maybe thats the yo-yo  bounce, like on old mates one in the vid a few pages back

[ajg193]

Haven't looked close enough yet as on my phone. How many repeats were done with each config? Really need 5 pulls to get statistically significant results, should also be done at random - ie not 5 in a row. That way you reduce the effects of systematic error

[ajg193]

@Roman can you integrate the curves to make xy plots of area under torque/hp vs runner length? Might be of interest

[ajg193]

Also might possibly be worth optimising ignition timing and possibly fueling for each configuration - more ve at resonance points will result in different flame propagation behaviour. 

 

[kpr]

Ive tired stuff in the past.    only very significant change in afr or timing has an effect.  timing is it the sweet spot,  few degrees either way no change.   afr;  11afr  right up to near 14afr same power.     done over 300 runs on this thing so know how it behaves . the biggest effect on power is oil temp.  and of course if tied to dyno different which it never moved off.

This plot below I went full retard and forgot i loaded the  normal tune back on at end of day.      soo  the afr was a lot richer on the day 2 blue run.    still only maybe 1kw change. oil temp makes up to 3kw change. 80 - 100c

 

[RUNAMUCK]

Awesome work guys. I can watch stuff like that all day.