Failed Experiments (FM Live)
HTML-код
- Опубликовано: 15 сен 2024
- This week on FM Live, Keith talks about product testing and experiments.... and how they sometimes just don't work out, while showing a simple test rig for bell mouth air inlets.
Original live video recorded 7-9-20. Join us every Thursday on Facebook for new live content!
Flyin' Miata is the world leader in Miata performance. We offer a wide range of parts for all years of the Mazda MX-5 Miata.
www.flyinmiata.com
970-464-5600
I'm an engineer. The best part about this video is Keith admitting he basically blew his entire week on the experiment because he was gonna win damnit. I have been there.
Watching you play with cardboard boxes and vacuums in your garage brings me great joy.
this is important, a failed experiment doesn't mean its a failure, just the iteration isn't the final one.
I wish we learned about this in high school science! Had to actually be in college for engineering to learn about stuff like this.
Yea, I have a feeling that scientific literacy would be so much higher if the education system found a way to demonstrate to kids how the underlying concepts impact the things kids actually like. Unfortunately no education system in the world manages to do this and it really is up to parents to do it (who are either incapable because they don't know the material, or don't have the time because... Duh, or both)
It's not even about paying teachers more either, to do the above you need a much much lower student to teacher ratio so the teacher can engage individually.
Like I can guarantee I can make a scientific or mathematical connection to literally any hobby a kid might have but I can't possibly do it for 30 kids in the span of an hour
To be fair to all the science teachers out there, this is really engineering. But definitely, practical experiments can really illustrate real-world applications. And seeing something as simple as a little cone affect the airflow through a tube brings it to life.
During a third year physics of music course in university, I had to do a lab session that involved finding the nodal points of a taut string by plucking it and stopping the string at the node - like playing harmonics on a guitar. Easy enough to do with math but the lab was intended to show it experimentally. The TA supervising the lab was blown away that you could actually hear octaves and fifths etc. It's fundamental and it's exactly how most instruments work, but this TA had never made the mental connection between the theory of a vibrating string and actual sound.
What's the saying ? I hear - and I forget. I see - and I understand. I do - and I remember.
@@FlyinMiataVideo Exactly. You can understand the concept without diving deep into the mathematics but without understanding the concept, the math is just some numbers and formulas. You really need both to apply it to designs but everyone can benefit from understanding the illustrated applications. Especially when the illustration is well done.
I would like to point out that this video came out ~3 years ago... which is how long it took me to finish anesthesia school. Knowing laminar flow and Reynolds number? Worth the wait.
Can't wait to blow money I'm finally making on perfecting my brake ducts.
Cheers!
This is going to sound crazy, but if you want to mess with the shape of the bell and not have to make it. Look up Bass Drum Port and I think you can come up with something to use! There is a company called Kick Port that makes them and a bunch of other companies as well and they come in a bunch of sizes too. Thanks for another good video!
I think the question of NB SC on an NC was very fitting. Unless the fabricator has cubic dollars or cubic time, it too will be a failed experiment.
So what was the impression on all the different set ups on the ND? It is still good for us to know, I just installed the koni/FM spring stage 1 kit. Would actually make a nice short video to talk about it.
The v1 Bilsteins (2016-17 soft top) were okay but not great with our springs. The v2 (all RF, 2018+ soft top) were not good with our springs as they don't have sufficient damping to control them. The Koni/FM setup was by far the most controllable.
We have in-car footage from the testing, we may revisit it and see if we can salvage something.
I am not an engineer but surely it doesn't help you have a massive hole in the cardboard box (the handle that's being propped open by a red hose) and no seal on the top of the box, those are MASSIVE leaks. There's no way you're going to get a proper reading on your airflow sensor when it's so much easier for air to go through the handle instead of trying to force the airflow sensor flap open. (The cross sectional diameter of that open handle flap is nearly the same size as the cross sectional diameter of your airflow meter)
Secondly isn't the entire concept kind of flawed ? A velocity stack (bell or bellhorn intake) is really only going to help when you have a decently large pressure differential between the intake and exhaust. Are unpowered brake vents/ducts ever going to create a strong enough pressure differential to REALLY extract much benefit from a velocity stack ?
I just want to clarify i'm not saying velocity stacks dont work, they clearly do. But how much marginal benefit will you get out of this compared to say, locating the end of the intake in a better location (to minimize kink's in the ducting, to access cooler air, to access air from a location to maximize the "Ram" effect) I'm really not sure ...
With that said, there's very little reason to NOT use a velocity stack, packaging wise you require just a BIT more room than if it wasn't there so if you have the space go for it. You should still always try to maximize the diameter of the ducting FIRST and the inlet shape should be a secondary consideration though ...
Dominic the handle in the box has blue tape covering it.
The handle hole is sealed with tape on the outside.
I agree that you want as much cross-section as possible, but anything bigger than 2.5" is difficult to package in a Miata. The Targa Miata uses 3" hose and the big hose limits wheel lock travel and the packaging at the rotor is very tight. You end up with some very tight bends and if you rub the wheels too many times, you get a hole. It's not a setup that would be acceptable on a lot of applications, so we're looking at how to optimize flow for a 2.5". If you're going to have brake ducts of any size, why not make them work as well as possible by simply choosing a better inlet design?
You've got a lot of pressure on the nose of the car. We've talked in the past about how to drop pressure at the other end. It all works together.
@@FlyinMiataVideo absolutely agree with you that the marginal gains are worth going after even if they are exactly that (marginal) as long as it doesn't take a tremendous amount of effort/cost to do.
I really do (and I'm sure others as well) appreciate you guys doing the testing and (furthermore) SHOWING the testing to us (your consumers).
Personally I'm kind of on the fence about chasing the marginal/incremental fraction of a second or gram or whatever it might be as with all hobbies (gold, watches, cars, guns, computers, whatever) there are lots of snake oil salesman out there and often times the bigger the claims the more dubious the actual results.
Personally I both love and hate objective stats... There is a time and a place for it (I don't think anyone needs to prove velocity stacks work... It's such a recognized and standard technology it's kinda like buying a 777 and asking the Boeing salesman to prove aerofoils/wings work...). As long as you're not trying to sell it as something that it isn't (claiming 200% increase in cooling for example) or charge an obscene amount for what is proven technology because it's "new" (apple trying to sell you a new feature that android has had for generations cough cough) I think consumers come out ahead.
Data is great but not everyone out there is a data scientist and bad data is often worse than no data. So much pseudo science out there nowadays that it's incredibly frustrating for anyone who's even a little bit scientifically literate to more help body murder at least once a day because of companies making impossible claims.
On a completely unrelated note, could you do a suspension alignment/geometry video (camber /caster/toe) ? I got your book (how to build a performance miata) but it (along with every other internet resource I can find) doesn't adequately explain how toe affects stability and tendency to understeer or oversteer (primarily at the front).
I always try to approach the problem by imagining the car going straight down a road and suddenly one wheel hits a patch of ice or oil slick ... For toe out at the rear wheels it's clear the wheel with traction will induce a moment on the car which turns in the direction which will cause weight transfer towards the grippy rear wheel, thereby this is a positive feedback mechanism and is unstable (toe out at the rear will make it easier to spin out). This is in agreement with real life. At the front though my intuition diverges from everything I hear from tuners and engineers. My logic is toe out at the front is stable, imagine same situation, I have toe out in the front and my left tire hits a patch of ice or oilslick. My right wheel because of my toe out will try to initiate a right turn which will cause lost transfer to my left wheel which is currently deficient of traction, with extra load, the left wheel will have more grip and try to initiate a left turn, bringing the car back to center... This however is wrong in practice as literally every resource I've seen says front wheels toe out is more twitchy and unstable because it creates some natural Ackerman geometry which while I agree with that point it only really provides an explanation for how easy it is to initiate a turn, not how stable something is (tendency to amplify/dampen random shocks from a state of equilibrium)
P.S. Sorry for the wall of text
@@2007MXV thanks
Didn't see it the first time.
requisite snarky comment:
it's 'N. A. C. A. ' each letter pronounced individually; not 'nahcah'... the way NASA is pronounced. That was all part of the marketing/mythology that came along with the space race and subverting the history of and/or Aero side of NASA.
or, at least, that's what my professors in college told me.
Good to know, thanks. The NACA also dates from a much earlier era, and the way of pronouncing acronyms may have shifted since it was formed.
I'm brainstorming a new intake for my 1.6, AFM delete and all, long story behind the idea, but would a bell mouth inlet for the intake provide any benefit?
Yes.
I see you 3d printed that Bell mouth... is that patented or would you share the file?
It's certainly not patented but we don't have the file in a sharable format. It's simply the cross section taken from the Blair article (available online) and rotated. It's a good introduction to 3D modeling if you want to learn!
I love this video, but I am wondering about the goal of the test. I think you've found a reasonable test for comparing inlets, but only as drawing from a static plenum. The ram air of a brake cooler is going to behave differently. I'm not suggesting that you need to go back to the positive pressure on the intake side because that won't make a bit of difference. I just think that an intake optimized for a static plenum might not necessarily be the best intake for a free stream. Besides that, while we know more air is better, that is somewhat difficult to quantify. You can work out the change in mass air flow from your test, but that will be difficult to correlate to the ram air. Even if you did that successfully, then taking that to a new convection coefficient would be extremely optimistic. Your uncertainties between what you've measured (pressure drop) and what you care about (temperature) are fairly large. In short, I think I'm saying you need a heated spinning plate inside a wind tunnel to develop a speed-dependent, variable-shape inlet so you can sell the most expensive brake cooling kit since the rockets that slowed the Curiosity rover's descent. No need to thank me.
Thanks Mark :) The amount of discussion this video has created is really fun.
We're not really looking at temperatures of the air - the amount of heating from compression, for example, is going to negligible. We're just trying to deliver as much air as possible to the brake rotor. What follows then is a separate area of study, but lots of air is a contributor. I'm not sure the ram air variant will work much differently as long as the direction of airflow doesn't mess with us. The air on the front of an air dam moves relatively slowly which is why it's high pressure in the first place.
We're likely going to have these on a race car in a few weeks with some documentation on actual rotor temperature. That's the true test, but difficult to perform during a live video shot in a home garage with a cellphone. This whole thing started as a "how do we demonstrate how these can have an effect?" and turned into a pretty fun couple of days.
link to print velocity stack ?
Sorry, it was a design we put together in-house for testing and was never published. We may not still have it.
@@FlyinMiataVideo por favor lo podes ver si lo tenes aun . es eun diseño tan perfecto . siempre busque algo asi para los itb . las ganancias comparado a un stack generico son del 1.5 -2 % . es mucho
please
I'm not a science teacher but I'm crying. you tested the bell mouth the opposite way than how you'll be using it. you tested it sucking air in through it. You'll be using it as a brake duct on the front of a car where you'll be pushing air through it. the very first setup with the bellmouth inside the box was the right idea for the application. But this test tells us zero about how it'd perform as a brake duct on the front of a car. If you choose to rerun this test the correct way can you please please please make anther video about it? Cause I've been seeing this bellmouth thing spouting about how its great for airflow, but all the testing I've ever seen has been only sucking air through it, and all the interpretation I've seen has been like "screw logic it worked here so it'll work in every conceivable application!!!". and furthermore, when testing pushing air through it, you should test with the bellmouth both protruding from the surface and flush mount with the surface. rant over, but please do the test.
not really
it's a static setup (unlike feynman's sprinkler)
all you care about is the pressure differential at the intake and exhaust ends
which is the same regardless of whether you have positive pressure at one end and atmospheric in the other vs atmospheric at one end and partial vacuum at the other
furthermore depending on how the exhaust of the brake duct is setup you might actually get a decent partial vacuum at the brake disc side if you could get a good enough seal at the rotor
Dominic Leung so prove it, do the test the other way too.
Dominic is correct, it's all about pressure differential. We're moving from an area of high pressure to an area of low pressure. It doesn't matter if that high pressure is higher or lower than ambient, the entire goal is to test the pressure ratio. It's the standard way of testing this sort of thing. It tells us a lot more than zero, it tells us that increases the efficiency of the inlet when air is moving through it.
That said, it wouldn't be that hard to do with the current setup. The shop vac has an outlet as well as an inlet.
The next step is to actually bolt a set of these to the front of a car and measure brake rotor temps. That's difficult to do in a live video shot in a garage, but it's the real test.
Flyin' Miata yes, I know your moving air from a high to low pressure area, that’s not the part I’m questioning. I’m questioning if you flipped the bell mouth around if it would still help things along.
Well, it would still be moving air from a high pressure area to a low pressure area. What's different? The absolute pressures, but that's not what drives the flow. As we approach supersonic flow speeds that absolute pressure is going to start to matter, but we're not anywhere close to that threshold.
I'm going to try it anyway simply because it would be easy to do and it will be interesting, but it's unlikely we'll shoot another video showing the same result.