Messing around with using a PC fan as car exhaust fume extractor. Fun experiment.

[AnonymousGuy]

So this deal started as a practical project where I wanted to be able to blow the exhaust from my car further away so it's less likely to get sucked back into the cabin when parked.  I put a fan on a tripod in front of the exhaust pipe...and that was good but then I decided I wanted to fuck around a bit.

 

So let's mess with solidworks' loft feature and 3d print something.  I made it sized for 140mm, 3mm thick wall, and the pipe portion exactly sized to plug into my exhaust pipe and long enough to clear the tip:

 

 

So CraftCloud, 7 days and $30 later (because it's better than running / learning my own printer): 

 

 

But here's the "fun problems":

 

2.25" is a pretty humongous restriction on this fan that's only being fed about 50cfm worth of exhaust at idle (it's a fixed number because the engine is pumping a known amount of air at a fixed load of idle). So it creates the weird effect of the fan doing "nothing" for the most part.

 

Air blows out like this...radially from the edges:

 

 

I want this: to achieve the goal of the exhaust being blown away from the car:

 

 

Soooo I dunno... .my intuition is that if I drill some holes in the shroud to introduce fresh air, it'll increase the mass of air the fan can work with.  More air mass would have more momentum to carry further?  Maybe the fan blades just work better when they're unrestricted?

[Alvin853]

Do you realize that car exhaust gases are REALLY hot? And 3D-printed stuff really does not like heat?

 

If you get car exhaust fumes in the cabin then most likely your exhaust manifold is leaking

[bmx6454]

if anything, it would make way more sense, and be far cheaper just to put a 3 foot section of universal exhaust pipe in it while parked(at least imo). i can't see plastic lasting anywhere near long term.. also worth noting, i never get exhaust fumes in my car while idling, even with all 4 windows open(at least not noticeable amounts). are you sure you don't have a leak somewhere from the headers to the cat? or is this parked in a garage or something while idling?

[Guest]

A combustion engine is an air-pump essentially.. it's going to overpower the fan at anything more than idle if it doesn't already.  As the last 2 people mentioned it sounds like you have an exhaust leak somewhere..

[AnonymousGuy]

20 minutes ago, Alvin853 said:

Do you realize that car exhaust gases are REALLY hot? And 3D-printed stuff really does not like heat?

 

If you get car exhaust fumes in the cabin then most likely your exhaust manifold is leaking

Yes I contemplated the temperature in the design, so I selected PETG for the purpose.  By the time the exhaust gas gets to the tip it's already been heatsinked by the rest of the pipe so while it's hot it's not melting hot.  Especially at idle where it's not a lot of gas to begin with.

 

If anything PLA would have been fine but I picked PETG since it's practically the same material cost.

 

(fan removed)

 

14 minutes ago, bmx6454 said:

if anything, it would make way more sense, and be far cheaper just to put a 3 foot section of universal exhaust pipe in it while parked(at least imo). i can't see plastic lasting anywhere near long term.. also worth noting, i never get exhaust fumes in my car while idling, even with all 4 windows open(at least not noticeable amounts). are you sure you don't have a leak somewhere from the headers to the cat? or is this parked in a garage or something while idling?

This is where I'm camping in the car and have fans pulling fresh air into the cabin so I don't have to deal with condensation on the windows and I can cook in there without building up more condensation.

 

EDIT: You do make a valid point about just running a section of pipe or hose to extend the tip away from the car.  It'd be viable but I was also trying to avoid carrying extra stuff since the cargo capacity gets maxed out when it's in camper mode.

[TeraSeraph]

You can improve the range that the fan blows the exhaust away by attaching a set of grid fins inside a duct on the back. I've made quite a few air filters and desk fans with it which easily triples or quadruples the distance that the flow extends out to. I also think you don't need the funnel and it would be more efficient to have a velocity stack which would pull more air in from around the pipe. You're not going to pull more exhaust out than the engine is outputting so the funnel is really just a restriction, sucking up the exhaust as it leaves the pipe will be just as effective.

[Blue4130]

The pressure of the exhaust is way higher than what the fan is capable of surely, isn't it? It's like trying to drink from a fire hose.

[TeraSeraph]

1 minute ago, Blue4130 said:

The pressure of the exhaust is way higher than what the fan is capable of surely, isn't it? It's like trying to drink from a fire hose.

It looks like OP is using an NF-A14 iPPC industrial 3000 which would probably keep up with a small engine or at the very least add some velocity to the exhaust flow.

[AnonymousGuy]

5 minutes ago, Blue4130 said:

The pressure of the exhaust is way higher than what the fan is capable of surely, isn't it? It's like trying to drink from a fire hose.

Not at idle.  At idle it's about 40-60cfm of exhaust (you know this based on the intake CFM which is needed for computing the fueling and it's basically 1:1 intake to exhaust...there's actually a bit more exhaust than intake because you've got combustion byproducts but it's something like 1.07:1).  And I'm also glossing over the volumetric efficiency of the engine changing based on RPM or some shit...I dunno it's what I read in about 15 minutes of google TBH

 

But yeah at even cruising loads it would exceed the fan.  That's why electric turbos and shit you find on ebay don't work.

 

13 minutes ago, TeraSeraph said:

You can improve the range that the fan blows the exhaust away by attaching a set of grid fins inside a duct on the back. I've made quite a few air filters and desk fans with it which easily triples or quadruples the distance that the flow extends out to. I also think you don't need the funnel and it would be more efficient to have a velocity stack which would pull more air in from around the pipe. You're not going to pull more exhaust out than the engine is outputting so the funnel is really just a restriction. 

 

I like all of these ideas.  I could tell immediately the fan wasn't spinning as fast as it should even not plugged into the exhaust pipe but trying to suck through the funnel on its own.

[TeraSeraph]

2 minutes ago, AnonymousGuy said:

 

But yeah at even cruising loads it would exceed the fan.  That's why electric turbos and shit you find on ebay don't work.

I want an electric turbo like those in the current generation of F1 cars. Never need to worry about turbo lag. 

2 minutes ago, AnonymousGuy said:

 

I like all of these ideas.

Best of luck! Let us know how it goes.

if you're really handy with 3D CAD you could make a stator for the fan which would be slightly more efficient than the grid fins but not worth the extra complexity in my experience. 

[Guest]

It would be much simpler to use some rubber exhaust hose like mechanics do to direct the exhaust fumes outside of their work area..

[AnonymousGuy]

6 minutes ago, TeraSeraph said:

if you're really handy with 3D CAD you could make a stator for the fan which would be slightly more efficient than the grid fins but not worth the extra complexity in my experience. 

Oh god I am not handy :).  I last really used Solidworks 10 years ago (don't think Fusion360 even existed yet) so half the point of this project was to just mess around and relearn how to do some stuff.  Especially nowadays where prototype manufacturing is cheap and easy.  I actually tried to learn Fusion360 but there was a bunch of things that it seemed like it didn't want to do at all (projecting extrude cuts) so I went back to Solidworks.  

 

[Whatisthis]

Yeah, just get you some rubber “condoms” you can put on the exhaust tips and fry that exhaust away from the car.

[kb5zue]

In southern California drivers have solved this problem by just adding a 3 foot pipe extension so that exhaust fumes are directed into the other lane.  This is especially true for trucks that have the added condition of "rolling coal" to blacken the windows of the vehicle in the lane next to them at the stop light.

[bmx6454]

56 minutes ago, AnonymousGuy said:

EDIT: You do make a valid point about just running a section of pipe or hose to extend the tip away from the car.  It'd be viable but I was also trying to avoid carrying extra stuff since the cargo capacity gets maxed out when it's in camper mode.

i could see that, but depending on it, it may take up less space than the current build. something like this which is flexable would be good if you could figure out a reusable way to slip it on without wearing out quickly :https://www.amazon.com/Walker-36322-Exhaust-Pipe-Flex/dp/B001AX9WJQ/ref=sr_1_4?keywords=flexible+exhaust+tubing&qid=1694562648&sr=8-4

edit: or even dryer vent tubing would be sufficient for the temperatures by the time it is at the end of exhaust.

 

either way, i'm not trying to knock your idea, i do think it's creative, just think it could be done simpler is all. i would also be concerned about possibly restricting airflow causing extra wear on the engine(not a lot, but it could add up over time, similar to when a cat is going bad).

[bmx6454]

45 minutes ago, Blue4130 said:

The pressure of the exhaust is way higher than what the fan is capable of surely, isn't it? It's like trying to drink from a fire hose.

at idle it is very low, i would be curious what the model of op's car is, so i can try and look up the expected amount of flow at idle.

[AnonymousGuy]

3 minutes ago, bmx6454 said:

at idle it is very low, i would be curious what the model of op's car is, so i can try and look up the expected amount of flow at idle.

3.5L 6 cylinder engine, 825rpm if you want to get fancy computing cylinder fill volumes and whatnot

 

I just went off the mass air flow reading that you get over the OBDII port where it was sitting around 40cfm.

[AnonymousGuy]

Here's the napkin theory of where I think this can be improved:

 

 

Uncork the fan by allowing it to draw fresh air from slots and then probably get a sort of venturi effect pulling the exhaust gas out.

[TeraSeraph]

1 minute ago, AnonymousGuy said:

Here's the napkin theory of where I think this can be improved:

 

 

Uncork the fan by allowing it to draw fresh air from slots and then probably get a sort of venturi effect pulling the exhaust gas out.

Can you print something that fits over the exhaust rather than sitting inside?

[RevGAM]

14 hours ago, AnonymousGuy said:

So this deal started as a practical project where I wanted to be able to blow the exhaust from my car further away so it's less likely to get sucked back into the cabin when parked.  I put a fan on a tripod in front of the exhaust pipe...and that was good but then I decided I wanted to fuck around a bit.

 

So let's mess with solidworks' loft feature and 3d print something.  I made it sized for 140mm, 3mm thick wall, and the pipe portion exactly sized to plug into my exhaust pipe and long enough to clear the tip:

 

 

So CraftCloud, 7 days and $30 later (because it's better than running / learning my own printer): 

 

 

But here's the "fun problems":

 

2.25" is a pretty humongous restriction on this fan that's only being fed about 50cfm worth of exhaust at idle (it's a fixed number because the engine is pumping a known amount of air at a fixed load of idle). So it creates the weird effect of the fan doing "nothing" for the most part.

 

Air blows out like this...radially from the edges:

 

 

I want this: to achieve the goal of the exhaust being blown away from the car:

 

 

Soooo I dunno... .my intuition is that if I drill some holes in the shroud to introduce fresh air, it'll increase the mass of air the fan can work with.  More air mass would have more momentum to carry further?  Maybe the fan blades just work better when they're unrestricted?

Why not put a grid of smooth material inside to encourage correct flow, like @TeraSeraphsaid?

13 hours ago, AnonymousGuy said:

I was also trying to avoid carrying extra stuff since the cargo capacity gets maxed out when it's in camper mode.

Uh, the fan and plastic piece are extra...;P

 

Edited September 13, 2023 by RevGAM
Added second part.

[AnonymousGuy]

5 hours ago, RevGAM said:

Why not put a grid of smooth material inside to encourage correct flow, like @TeraSeraphsaid?

Uh, the fan and plastic piece are extra...;P

 

Probably will.   I'm not sure though about the optimization of grid density vs. thickness of the grid to direct the airflow.  @TeraSeraph comment on what you've seen+

 

I suspect there's some sort of tradeoff curve of flow vs. direction like pressure vs. flow.  Maximum directional -> zero flow because you've effectively made a really dense and really thick radiator.

 

17 hours ago, TeraSeraph said:

Can you print something that fits over the exhaust rather than sitting inside?

Unfortunately not, since they've already done that by putting an "exhaust finishing tip" on the pipe.

 

This is version 2 where I can easily tape up the holes to adjust the amount of fresh air intake.  The math works out to the area of the holes is about equal or slightly bigger than the fan blade area:

 

 

I was going to do a bean shaped cutout but figured the bread slice is easier to tape with the straight edges.

[RevGAM]

2 hours ago, AnonymousGuy said:

Probably will.   I'm not sure though about the optimization of grid density vs. thickness of the grid to direct the airflow.  @TeraSeraph comment on what you've seen+

 

I suspect there's some sort of tradeoff curve of flow vs. direction like pressure vs. flow.  Maximum directional -> zero flow because you've effectively made a really dense and really thick radiator.

 

Unfortunately not, since they've already done that by putting an "exhaust finishing tip" on the pipe.

 

This is version 2 where I can easily tape up the holes to adjust the amount of fresh air intake.  The math works out to the area of the holes is about equal or slightly bigger than the fan blade area:

 

 

I was going to do a bean shaped cutout but figured the bread slice is easier to tape with the straight edges.

I really think you're overthinking the attachment. Still, if you go through with it, I'd like to know the outcome.

[TeraSeraph]

5 hours ago, AnonymousGuy said:

Probably will.   I'm not sure though about the optimization of grid density vs. thickness of the grid to direct the airflow.  @TeraSeraph comment on what you've seen+

 

I suspect there's some sort of tradeoff curve of flow vs. direction like pressure vs. flow.  Maximum directional -> zero flow because you've effectively made a really dense and really thick radiator.

 

Unfortunately not, since they've already done that by putting an "exhaust finishing tip" on the pipe.

 

This is version 2 where I can easily tape up the holes to adjust the amount of fresh air intake.  The math works out to the area of the holes is about equal or slightly bigger than the fan blade area:

 

 

I was going to do a bean shaped cutout but figured the bread slice is easier to tape with the straight edges.

I use 1.5" deep fins with 1.5" spacing which seems to work just fine. You could make them slightly more dense since you're printing them but I don't think it's necessary. 

[Lurking]

On 9/12/2023 at 6:15 PM, AnonymousGuy said:

Not at idle.  At idle it's about 40-60cfm of exhaust (you know this based on the intake CFM which is needed for computing the fueling and it's basically 1:1 intake to exhaust...there's actually a bit more exhaust than intake because you've got combustion byproducts but it's something like 1.07:1). 

You are ignoring expanded volume of heated air. You know, the concept THAT MAKES THE WHOLE THERMODYNAMIC ENGINE RUN TO BEGIN WITH. Hot exhaust flow is significantly higher than intake flow.

 

Why not just turn off the engine when parking? Or put a hose on. Or don't open the window. Your cabin air is taken in at the front. Maybe worry more about the car in front.

[AnonymousGuy]

9 hours ago, TeraSeraph said:

I use 1.5" deep fins with 1.5" spacing which seems to work just fine. You could make them slightly more dense since you're printing them but I don't think it's necessary. 

Sent this off for print.  1mm wall thickness on the grid which is as low as I can go on the design rules.  It should mate perfect to the funnel sandwiching the fan.

 

 

I'm excited to see how much of a difference it makes in terms of air guiding.