Solar "Hump" for aerodynamics for Class A RV

[eXodus]

Yup, hence why I said earlier the blockage ratio should be under 5% (and preferably under 3%).

I would not treat these simulations as anything more than a bit of light comic relief. Do them for fun, but don't expect any actionable intelligence. This really is an area where many years of expertise is required.

Just a entry point to learn a new tool.

I'm mounting the panels with a engineering safety factor of 6. So I use 6x the amount of fasteners as required per spec sheet for a 150mph installation. +Glue

So I'm pretty confident that it will not pull out of the roof decking material of the RV. How good the roof decking is mounted to the structure? Best guess?

I really try to do my due diligence here to building a solid structure and not break something or create a dangerous situation.

The simulation discussion here is really fun and I guess more to the fuel economy part related.

 

[wattmatters]

Is that your dog in your avatar?

He's the chief aero officer.

We went for a morning walk at our local beach (we have a designated off-leash dog beach) to check out some boundary layer dynamics.

Next time you're at the beach, take close note of the small patterns in the sand, especially leeward of any small protrusion. It can show a streamlined pattern, with the sand settling into a position where the wind is least energetic around the protrusion. Naturally forms a shape with low Cd.

 

[curiouscarbon]

dog

 

[AndyRonLI]

OK, so the bigger one just finished. It looks much better to me. Keep in mind that this shows air velocity and direction, not pressure, force, lift, or anything else. Also keep in mind that this represents less than one second of "reality." The RV instantly appears in the air, and it is already going 100 KPH. It is similar to what would happen if you were tailgating a semi and then suddenly (instantly) changed three or four lanes into clean air.

One thing I don't like so much is that the color map is adjusted at every frame. That makes it harder to compare the last frame to the first, for example. Anyway, here you go:

View attachment 79326

I'm currently running the "closed" model with a bigger air space (700x500). It is taking three seconds per frame, so it will be tomorrow before I can post it. Enjoy!

Sorry, that last one looks even wronger. More like an oblique shock off the front end, rather than a bow wave. Not sure what this solver is doing. Could just be way to early, but that looks pretty steady.

 

[AndyRonLI]

Should we start a dog thread? ;-)

 

[corn18]

Should we start a dog thread? ;-)

Yes! They always make me giggle. Cats, too.

 

[Cajunwolf]

Unless you're mounting a solar panel on your dog or cat, I would think it off-topic. I tried, but my cat didn't go for it; go figure.

 

[corn18]

Unless you're mounting a solar panel on your dog or cat, I would think it off-topic. I tried, but my cat didn't go for it; go figure.

Did you use a flexible panel?

 

[justgary]

Sorry, that last one looks even wronger. More like an oblique shock off the front end, rather than a bow wave. Not sure what this solver is doing. Could just be way to early, but that looks pretty steady.

I am pretty sure that the solver space was just too small, plus the RV did have those holes in it (which I assumed would not matter). The bigger ones finished last night, so here they are. The bigger one behaves much better, while the little one works until the flow falling off the top of the space fails to hold the rest of the flow down. In other words, the space isn't large enough to not impact the solution. It also helped to make sure that the front "tires" touch the ground, which wasn't the case before.

To answer the previous question, yes, the road is moving at the same speed as the air flow coming in from the left.

Note here that Flowsquare calculates delta time based partly on the size of the workspace, so these two simulations cover a different amount of time even though both are 30,000 frames long.

700X300:

700X500:

 

[wattmatters]

Should we start a dog thread? ;-)

Since the dog's name appears to be "Flo" that would seem to be on topic.

 

[curiouscarbon]

I am pretty sure that the solver space was just too small, plus the RV did have those holes in it (which I assumed would not matter). The bigger ones finished last night, so here they are. The bigger one behaves much better, while the little one works until the flow falling off the top of the space fails to hold the rest of the flow down. In other words, the space isn't large enough to not impact the solution. It also helped to make sure that the front "tires" touch the ground, which wasn't the case before.

To answer the previous question, yes, the road is moving at the same speed as the air flow coming in from the left.

Note here that Flowsquare calculates delta time based partly on the size of the workspace, so these two simulations cover a different amount of time even though both are 30,000 frames long.

700X300:

700X500:

thank you for using so much computer time to render! ?

as a total CFD (enthusiastic) amateur, the discussion about domain dimensions relative to simulated object very interesting, learning!

kindly may i ask for the project file? i have been having trouble getting my projects to work after so long of not using the software

 

[curiouscarbon]

Which is the more critical side? trailing or leading?

for Upforce, as a total aero amateur, i expect slight time averaged Up force on Trailing/Rear one
for Downforce, as total aero amateur, i expect slight time averaged Down force on Leading/Forward one

leading panel i expect slight down or up force. down from leading stream, up from air being shot up by inclined plane of upper half of windshield
trailing panel i expect slight up force from any air rolling around the left/right side and pushing up like this picture



as mentioned, like 90%+ of the friction is happening at the frontal face, perhaps suggesting only 10% of air force remaining to mess up the panels

excited about this project

if overbuilt with sika holding bolts and stuff, i would not expect unscrewing, nor would expect an acute stress failure, but every few weeks at first, then every few months, check up on the thing, maybe just get a ladder climb up and wiggle it with my hand to check for give. if there's a crack near the seams of connection or give, then maybe maintenance event time.

perhaps after installed it might cause new weird wind noises at speed in cabin. anything like flashing that creates a more smooth sweeping figure vs rough space for leading edge of panels might reduce this.

i'm CFD amateur hour! i am guessing! haha but i like physics and engineering.

 

[eXodus]

I am pretty sure that the solver space was just too small, plus the RV did have those holes in it (which I assumed would not matter). The bigger ones finished last night, so here they are. The bigger one behaves much better, while the little one works until the flow falling off the top of the space fails to hold the rest of the flow down. In other words, the space isn't large enough to not impact the solution. It also helped to make sure that the front "tires" touch the ground, which wasn't the case before.

To answer the previous question, yes, the road is moving at the same speed as the air flow coming in from the left.

Note here that Flowsquare calculates delta time based partly on the size of the workspace, so these two simulations cover a different amount of time even though both are 30,000 frames long.

700X300:

700X500:

that a great results. Thanks for taking the time and running them. The larger space seems to be more consistent. In the smaller spaces you got at some point where the flow detaches and just goes straight up to the sky.

In both simulation you also see those vortexes forming at the back - and I know I got those in real life - because road debris is getting drawn up to my rear window. So parts of the simulation are agreeing with real events.

Seems like the panels are directing the airflow up above the random items on the roof in a the first 10-15k steps.
Wonder how much that would be different with the model without the panels.

Would you mind running the model without the panels? Or upload the config file then I can run it.

Hope the weather is nice this weekend - I wanted to install a low profile Fan - to get the second panel of the Solar Hump installed - to start collecting some real measurements on that.

 

[justgary]

Seems like the panels are directing the airflow up above the random items on the roof in a the first 10-15k steps.

More likely the windshield doing it. Hiding the front edge of the panels a few inches back will probably take care of a lot of potential issues, as will blocking in the sides and rear so that air doesn't swirl behind in a major way.

Would you mind running the model without the panels? Or upload the config file then I can run it.

I'm running the RV_BASE model now at 700X500. I probably won't let it go the whole 30,000 steps if it looks like the front flow is stable. Here are the two files in case you want to run it as well:

 

[eXodus]

Got the low profile dometic vent installed. Should perfectly fit inside the second panel frame. The vent only vents to both sides, no lid to be raised. Very cool old product.

 

[Cajunwolf]

@eXodus LoL, Eternabond Tape, the Duct tape of the RV world. Only a fellow RV'er would notice that.

 

[eXodus]

@eXodus LoL, Eternabond Tape, the Duct tape of the RV world. Only a fellow RV'er would notice that.

I like the tape. I think it's easy to work with seals well, don't need to worry about a tube of lap sealant drying up when I don't use it.

Some manufacturers started using it as standard between the front and rear fiberglass caps as a standard installation. So must have some merit.

 

[curiouscarbon]

if it moves, but should not, eternabond
if it does not move, but it should, wd40

gradually figuring out how to use the new flowsquare; the config.txt file format seems to be a bit different. still learning how to operate it.. got a short term license for the new version flowsquare and it maxes out all the cpu cores on my computer when a project runs in simulation

 

[eXodus]

if overbuilt with sika holding bolts and stuff, i would not expect unscrewing, nor would expect an acute stress failure, but every few weeks at first, then every few months, check up on the thing, maybe just get a ladder climb up and wiggle it with my hand to check for give. if there's a crack near the seams of connection or give, then maybe maintenance event time.

I got a fixed mounted ladder. Further my RV goes on bumpy dirt roads, so I'm usually inspecting the items roof at least once a month. Probably with the new panels mounted I will be up there after each drive

I had a few screws not getting good grip (4 out of 24) in the roof substrate. They are not pulling out - but I can't get any torque on them.
So I'm going to wait until the Glue dries (without air in the screw hole - could be a couple of days) before I cover the screws- with Sealant or eternabond.

If the screws in question, still turning after the stuff sets - need remove them - fill the holes with Epoxy - get a larger screw and try again.
I used 1/4 roofing screws - the mounts permit up to 3/8.

 

[justgary]

For your continued entertainment value, here is the "RV_BASE" version, which has no panels. I think it is important to note that this represents only a glimpse of some sort of reality, and at the very most it is a static case down the centerline of an infinitely wide vehicle. Dynamic loads (e.g. a semi truck going the other way on a two-lane road) are not considered, and could have an enormous effect on your installation.

I would not screw into the roof material unless I knew I would also hit the structural member below it, but that's just me.

RV_BASE_NEW (no panels) at 100 KPH (62.2 MPH):

The vortex shedding at the tail (at roughly 33 Hz) kind of suggests that maybe the panels will help a bit since that one eventually stabilizes. I'm not sure the fidelity here is good enough to even say that, but it is a possibility. The tail here does get longer with each oscillation, so it may eventually stabilize but with a lower damping coefficient.