Mods for 90's Citroen Break (STW for folks in US))

[Piwoslaw]

Yeah, finding a good stretch of road isn't easy. Driving 20 or 30km just to your testing place will get everything warmed up. And I've found out that it's better to sacrifice more time and do more tests in one go, than to waste time and fuel on only a few tests done on different days. I used ~6 liters of fuel and 4-5 hours when testing my Kammback, but it was worth it.

[jtbo]

I did drove nearly 150km last evening/night, every place I remember being flat is not really flat, there are small up and down mini hills or then straight is not long enough.

I did found one place which gave me 50 seconds of coasting, still having small amount of uphill at both ends, but rather small I did though, did plenty of runs, put data to spreadsheet and I get negative cd, of course everytime there is downhill when I start the run, to both ways and uphill at the end.

For me it looks like that getting cd reliably measured it might cost so much in fuel that it would be rather pointless. Trying to find suitable road has already cost more than mods will cost and based on last summer's experience, I doubt that I can very easily save enough money to get even these costs covered with low mileage I do, so maybe it is best to forget testing and just apply aerodynamic aids and do hill experiment on known place at known weather and speed to estimate coasting abilities.

Wet road conditions change coasting dramatically, I'm barely managing to keep 75kph (starting speed 75kph) on one steep downhill, where I remember last summer speed being easily around 110kph with mods (starting speed 65kph), without mods and at dry weather I was above 90kph (starting speed 65kph).

With driving over 150km in search of good piece of road, I found out also 2 rabbits and 1 truck, nothing else moving out there.

Edit: I have been thinking a bit of Kammback, it will be quite small as there are no large space between end of roof and end of bumper and even it might be illegal to put any kind of kammback, it is at least less likely to be major offence when it is within body lines.

I have been reading Milliken's Race Car Vehicle Dynamics for other purposes, then I have been reading Aerohead's posts and generally researching this subject more.

I'm thinking this kind of bit agressively angled design, because of short distance and relatively slow speeds I use, but now I'm not sure if I should use flow separator at edge of Kamm to prevent vortex behind the car? I plan to fill back of car from air that runs from sides and under the car, but I'm afraid that without flow separator it might cause big vortex behind the car under the Kamm.

Red bit is one I wonder if I should put in there or not. I can of course make it detachable so that I can test with it and without it, but guessing before making it is always fun


Vekke posted link to pdf in Aero section where A2 Audi got good cd reduction from small straight pieces added to rear when they did lower angle of diffusor, that is something to think about too, but I doubt that I can attach lot to bumper, also tail light might get blocked when viewed from side and that is something to avoid.

Bellypan and wheel spoilers might be ones to adopt from that however and rear wheel arches I have already planned to block, front wheel arches would be great to block too, but it is not very easy to do well, however there is lot to be gained as I remember how at front even small tire arch gap reduction made noise a lot less.

[jtbo]

Some WIP photos from new part I started to work with, it does need quite a bit of work still, also that is not necessarily final angle as sides still need cutting and attaching to make angle right, but it is starting to form from there.





Rear diffuser angle is bit difficult one, I try to start with as close to zero degrees as possible, hopefully I can make it adjustable so that I can make changes to it relatively easy way.

I still haven't really thought much of how to make diffuser, but it might be same material as this Kamm, does not weight too much and when attached well should stay well on place. 0.5mm galvanized sheet metal, which means that I need to make some parts to make it bit less flexible for the Kamm at least, for diffuser, I think there is enough places to attach it so that it is not flapping etc.

[Piwoslaw]

Looks great! I'm sure you'll notice a difference
I have two suggestions:
- Use some reinforcement on the underside. I noticed that the center of the Kamm slightly lifts during the two times I've gone faster than 120 km/h, and tends to sag when wet snow starts to pile up.
- Make the corners a little more rounded. This will reduce any vortices that may come off of those points.

[jtbo]

Originally Posted by Piwoslaw

Looks great! I'm sure you'll notice a difference
I have two suggestions:
- Use some reinforcement on the underside. I noticed that the center of the Kamm slightly lifts during the two times I've gone faster than 120 km/h, and tends to sag when wet snow starts to pile up.
- Make the corners a little more rounded. This will reduce any vortices that may come off of those points.

Reinforcements and more rounded corners are already planned, also it needs to have something put to rear as sharp edge of sheetmetal is undesirable feature, but those are indeed currently what it needs + some cutting and shaping.

I had wooden Kamm last summer and even it had some reinforcements below it did start to move around 110kph.

Also it might need some braces to corners from center, below glass or opposite corners as sideways movement must be stopped too,even there probably won't be lot of it, still it would put more strain to attachment points to not have such supports. Alloy rivets are my choice of attachment as could not find anything else for that. (rivets are called pop rivets here, hard metal core and aluminium rivet, special pliers are used to attach, so it is one side job, easy to attach and relatively easy to take out too as one just needs to drill rivets out.)

I probably put also X-shape reinforcements, strips of sheet metal, 90 degree angle and drill holes + put rivets there. I could weld, but welding 0.5mm galvanized sheet metal is something I prefer not to learn at the moment, thick metal is no problem, but my welding machine is not too great with thin stuff.

Trunk lid is made from plastic, so it is not quite so strong as metal lid, but with this low weight it should handle well when I use enough rivets to share load over multiple attachment points.

Kamm is also designed so that I still can open trunk lid.

[jtbo]

I did another coast down test, as I was given different formula which does not need so long road.

However I have not yet had luck to get formula to work correctly, problem is my inputs to excel, I'm pretty sure.

Basicly there one starts at some speed, coast down to certain speed and marks up time it took, then second test at slower speed.

I did put those speed tests to spreadsheet from instructables, modified spreadsheet so that it takes only two first rows into account and that way I got results of cd = 0.3697 and Crr = 0.0187

Was there thread about coast down test values from this tool? It would be nice to compare if that is completely out of window result.

If I get that other formula to work in excel I will share spreadsheet and instructions as it would be nice to know differences between methods with bit larger sample base.

It is Bosch formula, so it should be somewhat usable I believe.

edit:
Here is the spreadsheet, it is not working too well for me, but perhaps my road was not quite level, but test with your data and post if it does work, it really should work just fine.
http://jtbo.pp.fi/images/ZX/science/cd_crr_calc.xls

Idea is to test how long it takes to slow down to 80 to 70kph test 2 is for slow speed, no aero influence so 40-30kph is ok I guess, slower the better probably. I have posted earlier link to spreadsheet that contains really lot of car models with their frontal area and cd value listed, it is at aero section of board in topic that contains cd information of vehicles, so one can try to get frontal area from there.

edit2: I learned that there was error in formula, one plus sign needed to be minus sign and it works now even for me, updated spreadsheet to reflect correction.

So my cd is 0.36-0.38, can't get much more accurate without taking more measures of measuring frontal area and better estimation of weight.
CRR is 0.012, but I need to test even lower speed to really get air resistance out of equation, testing from 20kph or so should be most accurate.

[jtbo]

I did estimations with tool found here:
http://ecomodder.com/forum/showthrea...ower-6341.html

I don't know how accurate that tool is and how accurate my engine efficiency and tire rolling resistance parameters were, but according to that tool, I should have had cd of 0.26 or so at last summer with aero mods or perhaps even less.

I had minikamm of crude design, rear wheel well skirts, partial front wheel well skirts, air dam bolted lower edge of front bumper, grille blocks, lower being poorly made.

No any kind of bellypan.

Driving was lights on, so maybe I should input overhead wattage there too (I did add now 150W), which means I should input even lower cd to get fuel consumption to level what I did record.

Stock vehicle has cd of 0.34-0.36, so having it down to 0.26 would be really lot, imo. Especially with quality of mods. What do you think, would it be possible with such mods to lower cd that much?

Link to my results

Results are bit higher, but I estimate some of the effect was hypermiling technique too, still there must have been big effect of aeromods as some trips did not contain hypermiling.

[jtbo]

I did add new fillup, 5.1l/100km, which is not so bad considering it was mostly short trips, also lower grill block seem not to matter much with those two gaps, engine is not staying warm, so might put upper grille block there too at least while weather warms up.

I made also coast down test from 20kph to 10kph, so that aerodynamics are not affecting much, this yield result CRR of 0.0098 which of course still contains some of the drivetrain friction, but clearly that low speed is important for testing.

As side effect cd did jump to above 0.4, it might be true as there is hole at other end of bumper, actually there is nothing left from bumper, which probably causes some drag as front of wheel and inner fender are completely exposed.

Next task is to put some covering there, if those sub zero temps just would cease.

I was attempting to gather some data from hills around here, but it was not quite a success, all did look fine but no log, also backup gps refused to work with pc, so I was left empty handed.

Must check if I have earlier data from that, it might be interesting to plot with speeds.

[jtbo]

I was thinking here about cooling of engine and how to remove hot air so that it would not be put under the car. One solution might be to make hole at front section of hood, just behind the radator, add small bump to front of hole, so that air would exit freely at low pressure are of hood and blend to airstream that runs over the car as is done with Lancer Evolution rally cars.

Also I was thinking that with such I could get away with relatively small openings at front, but that would compromise air route to IC that runs in hood center line. I would need then to make some kind of ducting or piping that would bring air to IC from lower part of front as I hope to put upper grille completely shut.

Radiator gets on way, maybe piping to go under the radiator as sides and top are impossible, but then it would cause piping to be on way of belly pan from bumper to engine bay which I have planned.

So do I make just hole to front of hood and scoop to rear/middle of hood for IC?
I don't know what is effect of scoops to aerodynamics, might be horrible...

[jtbo]

New pic:


Ok, pic is horrible, one can't see how it has actually smooth transition from side to top and how sides are pulled inside at least 15 degrees, top is perhaps around 9 degrees, I did estimate measure 5cm drop in 50cm distance, so that is 10% and if 90 degrees is 100% drop, then 10% must be 9 degrees, but that can be awfully wrong thinking, I'm not good with angles and measuring, so I did build whole thing by eyeballing it.

So with my latest coast down test, I had ended up to cd of 0.429, which can be possible, because end of front bumper is missing so from bumper air meets inner fender and tire, also no spare tire means that rear bumper is like a parachute.

Same road, but this time road was bit wet, which might cause poorer results than with same conditions as before, one pass to both directions and I got result cd 0.308 which might be bit too good, or maybe not, if I calculate right that is 28.2% improvement, but there is also now upper grille block in place too.

Subjective estimating is that wind does not grab car so much, car is also a lot more quiet, not much wind noise anymore.

I can coast down on places where I could not before.

I did make new fillup even I had not driven much, only 104km driven, from those around 50km with kamm. 4.56l/100km which includes quite many cold starts too. There is difference of 0.54l/100km and if that is because of kamm, then it would reduce consumption by 1l/100km, but that is as good method as lottery for estimating anything.

I always fill up so that diesel stays on ring that is at filler neck, also I put fuel in very slow and use same meter and same position of car, as I attempt to reduce errors from filling, but it is not still 100% that I get similar way of fillup each time.

Upper grille block is also affecting that result, so is ram air pickup, which is perhaps more of cold air pickup, but at least it is under the bumper at the air stream, whatever good that will do.

Upper grille block seem to lift snow up so that it does not hit on windshield anymore, also rear window seems to stay bit longer clear than before, water drops seems not to be removed from kamm quite easily, after 20km test drive I still had them there, whatever that tells.

In rally ZX kit car, there seem to be similar plastic pieces at front of wheels as new bmw/mercedes cars have, I believe that if I can copy location, size and shape of those, then I might get some gains, when I put rear belly pan to car as air guided underside the car is at the moment causing lot of drag.

My aim is to get Opel Calibra level cd out from this car, but I don't know if that is realistic at all, but I can always try, also reliably in 3.5l/100km level fuel consumption or less in this hilly area is what I hope, maybe hitting as low as 3.2l/100km if weather is good and level roads, but that might be impossible.

Best thing so far is that reduced wind noise.