Very true on variations of velocity depending upon where you are observing the anemometers velocity values.

Typically you "grid" off the circumference (or profile; if a square or rectangle or whatever).
The size of the cells in the grid should approximate the diameter of your anemometers impeller.
Average out the cell values for a better CFM or Velocity approximation.

Here is a decent how to.

However, 176 cfm is still quite a bit below what a 12" surface mounted Spal type fan produces. Karl still has to do some final testing of the output on the back side.

That's an excellent pile of ref info Charlie!

CFM Solution

176 CFM doesn’t cut it---is meager air. Solution = add second turbine near bucket. Provided around 50 mph of air flow at entrance to bucket, this puts CFM in 384 range. Now include low temp of air from ahead of rear wheel being pushed through condenser… concept has wings again. (Had no input concerning CFM been provided, system would be living in delusional state and info published would be wrong---I don’t support either, particularly the latter. Once again CG and Jonny… glad you chimed in on miscalculation. TY.)

Is adding second brushless turbine a thriller? Yes & no… is good for air movement but more hardware = more complexity. To get a reasonably good flow of air to condenser... this simply has to be. Challenge now = add no more than is necessary---am admittedly not always good about this, excuse… exploration-has-value. Having no commercial interest to account to, this project is able to explore with almost no boundaries. (A limitation being goal to finish---a.k.a. charge AC system---before summer’s heat is gone---2017 summer.)

Main mechanical challenge

= Space. Somehow, duct where I avoided positioning first turbine… that area will have to accept second unit. (Reason for not putting #1 turbine near bucket was curve in ducting.) Will have to minimize length of #2 turbine’s housing as a means to an end here. And, rather than stick turbine’s house inside 4” duct house, plan is to use 4-to-2.75” reducer/coupling. Did not do this for first version thinking house-in-house with insulation wrapping outer housing… that config would serve noise reduction. It doesn't. Having tested system, insulation-for-noise-from-duct doesn't matter---air/turbine-noise exists at both open ends of system. My take = wrapping ducts in thin sheet lead might have greatest impact on duct noise---and I looked into material cost & weight---but adding lead to car… not happening. Treme brought an "impregnated" noise barrier to light. Interesting stuff. Again, system being open at both ends moots value of insulating exterior of duct (for noise reduction IMHO.) Stack’s noise-insulation... that's turned into a beneficial cushion-against-car for stack. Insulation that will come will serve thermal purpose---goal being to minimize engine heat from being absorbed by cool air making passage from ahead of rear wheel to condenser behind it. (Alternative to this whole direction = condenser designed to fit ahead of rear wheel. Optimizing cool air flow---source>volume>speed>exhaust deposit---remains a challenge for that config as well.)

Electrical

System currently looks like this (below). Rather than add second Electronic Speed Control (ESC) and second 4-wire cable for second brushless motor, would be nice to feed power to both turbine motors from same ESC and cable---idea being to split service wires near motors. Doing this is purely theoretical at this moment---I’ve no idea if splitting is possible with pulse signals that drive brushless motors. I've also no sure idea if Arduino would serve two ESCs but believe that's doable. Have to look into these areas. Another question is amps... ESC is 40 amp unit---will that support 2 motors? Also, if brushless will accept a split service source, will 16awg wire handle current for both motors? If someone wants to school this electromoroon in this arena… am all for it!

http://forums.pelicanparts.com/uploa...1504366627.jpg

384 cfm may not be enough; you'd have to check what the backside velocity is.

Objectively you are trying to match or beat the efficiency of a 850 cfm direct mount fan.

I would not be so concerned about pulling air in behind the LH rear tire provided it is on the fender's outer side, alike the Kuehl design; we did lots of testing relative to header convection and cat radiance when in the early design stage 18 yrs ago and naturally for the Willy.
Convection gives you a slight rise in temps when idling if you have wind gusts from back to front, however the airflow of the design (Kuehl) moves the air outside to inside and down with baffling. Cat wise, the angle and height of the condenser above the lower valance supports avoids much of the cat's radiance.

See what you can play with in a 2 stage blower or turbine provided you got the amps to drive it.

Again, finish it, make it simple, and see what your flow is out the back of the condenser.

^^^ No question in my mind you went through thorough R&D on rear fender condenser positioning Charlie. From get go, and I'll repeat it here, with your system and referencing your "Ice Man" project, my '80 911 AC achieved 33 dF at the vent with 50 dF delta. Want to make it clear that no toes are being stepped on here---IOW, rear condenser with Spal works perfectly well to drive 33 dF out 911's vents. (Mess with TXV and/or thermo cap and below 32 dF is likely. RonO can attest to that. George, he makes ice in his 911's evap here in Miami.)

We're akin to climbing the mountain here---makes no sense except to those that like doing so. IOW, I build because... it's there---in my mind's-eye it's there. It's just damm amusing to explore ideas & stuff. And share the journey so it can be a mutual sort of mountain-climb for those of similar insanity. When I'm wrong, (you're the first to point it out! ;) and appreciatingly so I'll add,) I say it and move the fook on!

Am no longer looking to beat 850 cfm fan. Though I had!... but bear got me on that one. Agreed, air flow/volume through condenso is important. It must be that temp of air passing through condenso is also important. IOW, blow a tremendous volume of 250 dF air through a condenser then blow a small amount of 20 dF air through and end result is---for example---the same. This air-temp difference is what I'm playing with. Was from the beginning. Got lost in air delivery which was fun. CFM game is now what it is with #2 turbine/brushless added. There'll be no #3 for sure---summer's almost over (and I want that summer heat to charge system in!)

Am I going to beat 33 dF at my vent? OK... 32 is new goal---I have no interest in frozen evap. It is possible to adjust Arduino program to check ice-making capabilities but not inclined to go there. John, a.k.a. DrJ, did an amazing creative dance to get digital sensing & AC op program working---am not F'ing with it. If there's an "excuse" needed to climb this mountain... am using delta as justification. We're after more delta.

Again... no fookin toe steppin happein here. You know this 911ACchit better than ANYONE. By comparrison, I'm loose monkey with Dremel and some duct (following creative orders handed down from the WildBoars. :D )

Running two brushless motors off one ESC

First research info found was conflicting---some claimed it being a death march for motors and ESC. Others held it possible so long as motors were matched, ESC amp rating took combo of two motors into consideration, and load was even. If load went uneven, it was suggested that back EMF as read by ESC from motors would cause system caos. That was intel from a few BM forums. YouTube shed more light with visual demo---it does work with matched motors and ESC amp rating taking both motors into consid and---the important part now---regardless of individual motor load---which is important in this case.

https://www.youtube.com/watch?v=O0iMQ9rQNuo

This ref does not guarantee motors and ESC in hand will perform same as YouTube test but it’s good info to start from. I’ll bench motors together when #2 is received.

Load consideration

Load is important factor in this case as #2 motor (in sequence) is likely to operate with less load compared to motor ahead of it.

Divergent pitch

Looked at vid that references “divergent pitch” I found when using flexible/rubber hose in system. As vid motor starts, there’s a low pitch. As speed increases, there’s divergence. That high pitch following divergence is what I refer to as “annoyable.” When using metal duct, that annoyable pitch does not exist. Explained this intel to GF… she looked at me with---I’m certain---great admiration for intel she just gained.

https://www.youtube.com/watch?v=pXDNMv9t990


Misc entertainment:

Hand made jet engine. May not be headed to MOMA but thing actually works... (or does it? A few dubious elements on second consideration.)
https://www.youtube.com/watch?v=hEy6t5uWg-Y

…Running:
https://www.youtube.com/watch?v=jje0wV4l0tM

That diaper deal made a big impression on me because HOAs are very prevalent in TexASS, and in my opinion, are nothing more than legalized criminal organizations that allow small-endowed individuals to exert power over others . . . in other words, they SUCK!!! :)

Really enjoying the thread, KinkyK, and am learning a lot, bro!

Ya... HOA here is like government---lot's of talking & renaming streets, little genuine problem solving on impressive scale. Perhaps I'm too harsh. Perhaps not. Glad you're into this expedition RonO. Hope you're well away from challenges in TX Bro---that's a rough ride for some.

Looking forward to your Sunday muzac pic of the week. Something enchanting as usual... yes?! :)

Yes indeed - this wild boar is all over it! :D

Hurricane did not affect the DFW area, other than light rain and lowered temps for a couple of days (back to superhot, now) - terrible stuff a few hours south, for sure!

At your service, KinkyK, with this most "enchanting" offering of a very cool, Isle of Man TT vid that someone made to the song "Faster" by the Dutch, symphonic metal band "Within Temptation". I think it fits nicely with your quest for optimum airspeed! :D

<iframe width="854" height="480" src="https://www.youtube.com/embed/L0M_a8S4Y4c" frameborder="0" allowfullscreen></iframe>


And this is in keeping with the TexAss/Houston theme you mentioned above - old school Pantera.

<iframe width="854" height="480" src="https://www.youtube.com/embed/i97OkCXwotE" frameborder="0" allowfullscreen></iframe>

HC... very enlightening input. This visually explains delta in air speed test results found between "gummy" ducts (as in rubber or similar) and smooth metal duct. Delta over 3' duct = 2.8 mph with 26.5 mph being the higher metal duct's end/outlet speed.

...Friction over 3' duct length resulted in drop from turbine's static output speed of 49+ mph to between 23.7 (gummy duct) and 26.5 mph (metal duct.)

Fact that any calculation based on 100% volume-flow through a duct fails to consider wall friction... you've added a high-end perspective to air flow HC---much credit to ya.

Ahhhhhh.... damm fine selection RonO. Especially enjoy TT... soundtrack and race visuals... yeah---marvelously enchanting Bro! (Good to hear you escaped having to deal with stormstuff.)

Status

Am waiting on turbine blower #2. (2 more are also ordered as spares.) Tomorrow will install cradle mount for intake stack. Thinking through duct… plan is to try a hybrid: metal-with-minimal CEET. CEET being applied only to area just above wheel where duct is made into oval to fit into crevice between fender and car’s tub. Interior wall of CEET needs idea for smoothing it out---thanks to HColes for duct visual that illustrates air movement within duct. Idea to fit #2 turbine just ahead of air-bucket remains. #2 will have to be fit with parts in car in order to sort out curve of duct-work from bucket to turbine #1. Will bench test system before going to car to check some new unknowns (given addition of #2 turbine.)

Vid (not mine): Volts & amps for HP c7000 Blade Server system cooling fan unit being used. Am planning to keep existing 40 amp ESC based on what's seen here unless there's opinion against it.

<iframe width="373" height="210" src="https://www.youtube.com/embed/Crmz4-iWGvg" frameborder="0" allowfullscreen></iframe>


HPc7000 Blade System. Focus of vid is server system---skip to 11:17 to see cooling modules from where turbine fan here comes from---each module has a handle. (Handle rotates a few degrees in order to remove module.) 5 cooling modules are installed here. Noise level in vid is likely not with all 5 cooling units going full tilt… my guess is they’re operating around @20% here (provided all sound is from ambient mic.)

<iframe width="373" height="210" src="https://www.youtube.com/embed/s07wE2qz7Ww" frameborder="0" allowfullscreen></iframe>


840W blower---IBM Blade Server.

<iframe width="373" height="210" src="https://www.youtube.com/embed/pS8h8z1c0bI" frameborder="0" allowfullscreen></iframe>


Have questioned developer of 101mm (4" diameter) EDF brushless turbine concerning use of their "big-air" flight turbine in a "dirty" ground-oriented system (where only coarse-air filter exists)---will it survive injesting grit stirred up from pavement? Point is to continue research into ways of maximizing air delivery to bucket. If developer answers favorably, a good but potentially delaying wrench is thrown into gears-of-progress. Goal is to get AC system charged & running within 30 days.

http://forums.pelicanparts.com/uploa...1504667609.JPG

Stack's prepared---with Coastr's air-valve (TY Coastr)---to mount cradle-bracket and…


http://forums.pelicanparts.com/uploa...1504667609.JPG

...air-valve’s intlet's up against tub. Quick solution = pointing inlet up or down…


http://forums.pelicanparts.com/uploa...1504667609.JPG

= Debris catcher.


http://forums.pelicanparts.com/uploa...1504667609.JPG

Points in same direction as main intake a few inches below. Doesn't make sense IMHO.


http://forums.pelicanparts.com/uploa...1504667609.jpg

Distance from inlet edge of valve’s tube to end of useable horizontal space = 6 ¼”. Short cobra fits. See: intakehoses.com. Not thrilled to wait for more parts now or pizz away $s overnighting.


http://forums.pelicanparts.com/uploa...1504668066.JPG

Adding short section of tube... can do right away. This moves 90 backwards putting inlet in area of inward flare of wheel well. (Can see inward flare in wheel pics above.) Will get cobra if this doesn't fly. Plan: get this and bracket done tomorrow.

Volumetric air flow rate is not, in general, easy to measure. Of course it depends on how accurate you want to be. If you are curious about measuring air flow rates you can consider building what we call a wind tunnel with a counter blower. It would not be too hard to construct. However you need a couple of "official" orifices and a couple of manometers. The air flow rate of your complete system can be measured including the restriction on the fan or blower. It also helps to have a way to get the fan rpm, then you can see exactly where you are on the fan curve and closeness to the "knee" of the curve.

Am tempted to dive in to manometering deep end. Would be interesting to see how air flow is effected by various surfaces---in this case CEET and flat metal surface ducts. Of course is common knowledge that a smoother surface is more slippery. What I never visualized---beyond a turbulent progressing cyclone barreling through duct---was center part of that barrel moving faster than outsides. To measure speeds within this barrel is Frankenstienscience---if I only had the time!

Bottom line, what's coming out azz end is what counts.

Would like to incorporate some "slower" CEET duct into system. To speed CEET up AMAP, am thinking an interior surface treatment is called for. Is not so much micronic mph gained in this short-length case but sorting out what material is answer.

You seem educated on matters of flow. Am wondering if you have any comment on air flowing through CEET and rubber hose producing a high (annoying) pitch whereas smooth metal duct did not produce this noise. Same length of duct and same turbine speed for each test.

You can likely find papers exploring air flow rates and surface roughness. I doubt there is much flow rate to be gained in your case going from a "smooth" surface to a super smooth surface. One rule of thumb is to not make tight turns. I've now forgotten what you were trying to accomplish, I'd say put things together and see how it works. Your construction skills and energy are impressive. To measure air flow in a particular case (limited air flow range) there are other devices you can use other than the full blown air tunnel. Were you able to get the performance curve for that Hewlett Packard fan? I probably had one at one point. HP did a lot of research years ago on fans (after seeing ducted fans on RC planes) and got some fan vendors to play along to develop better fans for servers. The result is one of the fans you are using.

Interior surface of CEET (and similar rubber hose) has a gummy/sticky feel. Am assuming that air "feels" this gummy friction just as it's sensed by touching surface. Agree that exploring methods for improving air-flow in short section of CEET now planned is meaningless. Idea is to explore smoothing solution for other possible applications---is reason for various side trips (I take.) Also agreed on bends... While not in exact config, this (below) is very close... (this is prior to air-valve addition in stack. CEET section will likely be half of what's seen here---so approx 1.5'. )

http://forums.pelicanparts.com/uploa...1504879589.JPG

Goal is maximum air delivery to bucket/condenser with acceptable noise level. Curious find was high-pitch noise existing when CEET/rubber hose was used. Am hoping shorter and possibly interior-coated CEET will not serve up a high-pitch. Did not find docs on HP's server cooling modules. I've been in a few server rooms and always thought it was computers making racket---now I know is all the air being moved for server cooling! Evolution here too is from RC ducted fans (EDF's) to the HP unit---thinking being EDF turbine blades are too delicate for this app.

I may have missed this in the earlier posts, but how is this not best achieved by slapping the largest, strongest fan you can buy right to it? Seems like with fans and airflow, diameter means a lot more than RPM for CFM and noise. These things are capable of cooling 6L V8s...I'd be surprised if they wouldn't be enough to keep any AC condenser at near ambient.

A general rule - small fans make more noise compared to larger fans moving the same cfm. Are you pushing air across the heat exchanger (hex) or pulling through it? Pulling tends to equalize pressure across the face of the hex providing a more equal plenum effect. You don't have a big plenum. When pushing it is easy to create impingement areas (good) and also shadow and low pressure areas (bad). You are already doing very good compared to the original design that has recirculation issues.