Logo
Builders' Plan Gallery  |  Hip Pocket Web Site  |  Contact Forum Admin  |  Contact Global Moderator
November 24, 2017, 07:41:00 AM *
Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with email, password and session length
 
Home Help Search Login Register
Pages: [1]   Go Down
Print
Author Topic: EF-1 Speed Secrets  (Read 288 times)
0 Members and 1 Guest are viewing this topic.
Konrad
Platinum Member
******

Kudos: 29
Offline Offline

United States United States

Posts: 1,662


Topic starter
Measure twice cut once



Ignore
« on: June 28, 2017, 12:08:47 PM »

I think I should but this disclaimer up front: In EF-1 racing I have never placed in the top three with more that three participants in a race!* I attribute this to the caliber of the fliers I race against. Others like to point to my flying skills.

With that damning critique out of the way lets see what I’ve learned in my first year of flying EF-1 pylon. For those that aren’t aware of the USA NMPRA EF-1 Pylon class of racing it is a highly spec’d entry level 4 cell racing class with models that are loosely based on the full size Reno F-1 sport class of pylon racer.

Contrary to what the title might imply there are NO SPEED SECRETS! Well, other than practice practice!

First up is the control throws. While from the spectator’s view point pylon racers look very responsive, to the pilot they need to be set up as rather boring sport planes! This is key, the model should not jump around like a 3D model.

The ailerons should be set up so that it takes over 2 seconds to perform a 360 degree roll. Actually it should take the time the model travels from pylon one to pylon two. The reasoning is that if you have lost the “line” and need more than this amount of throw to come back on the course you have already lost the heat!

The rudder throw should be set up that it takes 7/8 right rudder stick to keep the model heading straight on the take off roll. While it is true that these model only need to be on the ground for less than 4 meter. You don’t want to be zig zag all over the place. With all the airplanes having very close to the same speed you can’t make up for a bad launch. Most races are won at the start. You want a straight take off with shallow climb out to pylon one. You want that extra 1/8 rudder stick to deal with any cross wind from the left.

Now setting up the elevator is a bit involved as it is highly dependent on the placement of the center of gravity! The center of gravity shown on most if not all plans (manuals) is far too nose heavy. This is done to give a new out of trim model a chance to survive the maiden flight. During the maiden flight after you get the model to trim level and are somewhat comfortable with the handling of the model you will need to test for the center of gravity.  From level flight roll the model up on the left wing tip (no rudder knife edge). With the wings vertical try to notice if the model is drifting to the landing gear (right) or the canopy (left). Most likely the model will be pulling real hard towards the canopy. This is an indication that you are nose heavy. You are carrying up elevator trim to compensate for the nose weight. You probably also noticed that you diving towards the ground!
Bring the plane back to level and land her. Now before the next trim flight to move the battery back 5 mm make whatever other adjustment the maiden flight indicated you need to make. Pay attention to the responsiveness of the elevator. Since you will be moving the Center of Gravity (CofG) aft the next flight will have a more responsive elevator.   If you could barely control the elevator with the manual CofG? Moving the CofG aft without cutting down the elevator throw might not work out too well. Perform the non-rudder left wing down knife edge flight  and notice that the plane should not be pulling as hard towards the canopy. Repeat  and keep slowly moving the battery aft until the model does NOT change heading in the left wing down knife edge flight.  This is close to the proper center of gravity for pylon racing.

You might have noticed that you are now using about 3/4 as much elevator as you did on the maiden flight. Now to set the elevator throw get about three crashes high and with the model going full speed roll up on to a left wing down knife edge and pull full elevator. If the model comes around nicely you don’t have enough elevator. If the model snaps you have to much elevator. Adjust the elevator throw so that from a high speed straight line flight into a hard full elevator stick turn the model just starts to protest (is on the verge of a snap). Note that in a straight line the model is flying faster than it will be on the race course, so you shouldn’t be near this stall point while racing.

Readjust your linkages so that you are able to maintain these surface deflections while using 80% of the servo throw. I like to use low rates (about 80% to 90%) for racing I then switch to high rates to help with the landings. This way I’m able to get the most from my servo’s resolution and power.

Oh, I should say calibrate your ESC for full throttle!

This is just a start, more to come.

All the best,
Konrad

*In the 80's I flew a lot of FAI F3D Pylon and AMA Quarter Midgets. To date I've only flown 3 EF-1 races so low standings really shouldn't be a surprise to anyone.
Logged

Cut it twice and it's still too short!
Konrad
Platinum Member
******

Kudos: 29
Offline Offline

United States United States

Posts: 1,662


Topic starter
Measure twice cut once



Ignore
« Reply #1 on: June 30, 2017, 12:54:04 PM »

Now what follows is rather difficult to describe, that is the acceleration induced roll. Often time called lateral balance.  If one wing is heavier than the other as you pull high G’s this apparent weight difference between the two wing increases  but the aerodynamic forces stay closely in balance (assuming both wings have the same twists and airfoils). With the model trimmed for level flight I test for lateral balance trim by making a sharp vertical pull up, not so sharp as to stall a wing, the model will roll to the side that is heavy.  

Rudder and motor side thrust, again trim for hands off level flight. Then gently pull up into a 45° climb for about 150 meter. If model yaws one way or the other adjust rudder for a straight climb.
 
Now in a pylon turn you don’t want the wings to roll nor do you want the fuselage to yaw. From the above mentioned trim condition pull a sharp knife edge turn (pylon turns). If the model whips back at you like it was hit with a tennis racket you are in trim. Most likely with EF-1 racers you will have noticed that the nose dropped. A lot of this has to do with the fact that the EF-1 racer is a bit under powered for its weight in that it can’t hold a level knife edge flight for long.  To help keep the nose up you will want to add more right thrust or right rudder to your trim set up. I usually add more right thrust.  This also helps with the take off roll. (recall that I said I often need over 6mm of right thrust as measured at the prop tips). Adding right rudder also is helpful but watch for tail wagging in yaw if you added too much rudder trim.

This should get you into a rough race trim. You will have to play a bit with all these setting as there is a lot of cross talk between the adjustments. For example aileron rigging can give the same results as the CofG in knife edge flight.

To give you an idea of how much work trimming is, it often takes me over 10 trim flight before I’m ready to race. So be patient!  Don’t think you can go to the races with a new bird and find your racing line the first time out.

Next time I’ll try to cover equipment a bit. The great thing about HIP is that I’m NOT constrained by concerns for their advertiser base so I’ll try to name names!

All the best,

Konrad
Logged

Cut it twice and it's still too short!
JohnOSullivan
Silver Member
****

Kudos: 4
Offline Offline

Canada Canada

Posts: 178


WWW

Ignore
« Reply #2 on: June 30, 2017, 03:48:40 PM »

Konrad:
For those of us in the rest of the world who have never heard of EF-1 would you please post rules or a link to the rules so we may have some idea of what in the world you are referring to.
Logged

John O'Sullivan
MAAC 5401
MACI 26
Konrad
Platinum Member
******

Kudos: 29
Offline Offline

United States United States

Posts: 1,662


Topic starter
Measure twice cut once



Ignore
« Reply #3 on: June 30, 2017, 04:34:18 PM »

John,
I'm sorry I thought I had stated what EF-1 was.  NMPRA EF-1 Pylon class of racing is a highly spec’d entry level 4 cell racing class with models that are loosely based on the full size Reno F-1 sport class of pylon racer, flown mainly in the USA


The current set of rules
http://www.nmpra.net/rules/EF1NewRules4_15_2014.pdf

All the best,
Konrad
Logged

Cut it twice and it's still too short!
JohnOSullivan
Silver Member
****

Kudos: 4
Offline Offline

Canada Canada

Posts: 178


WWW

Ignore
« Reply #4 on: June 30, 2017, 05:15:18 PM »

Thanks Konrad:
That puts everything in perspective.
Good luck in your cooling endeavours
Logged

John O'Sullivan
MAAC 5401
MACI 26
Konrad
Platinum Member
******

Kudos: 29
Offline Offline

United States United States

Posts: 1,662


Topic starter
Measure twice cut once



Ignore
« Reply #5 on: June 30, 2017, 06:04:58 PM »

Thank you John,
Actually the cooling has been working great, based on telemetry.  Now I need to find a way to make it lower my heat times.
http://www.hippocketaeronautics.com/hpa_forum/index.php?topic=22200.0

I've been performing the pressure  cowl mod for many of the "fast guys" I fly against. None are reporting any changes in the performance of their ships, but they seem to want the modification.

All the best,
Konrad
Logged

Cut it twice and it's still too short!
Konrad
Platinum Member
******

Kudos: 29
Offline Offline

United States United States

Posts: 1,662


Topic starter
Measure twice cut once



Ignore
« Reply #6 on: July 01, 2017, 01:17:34 PM »

Now I’d like to touch on the airframes. To save those that don’t want to struggle through my writing. I’ll start by saying that the best EF-1 ARF airframe is the Great Planes, Proud Bird. It is good enough to win the 2016 USA NATS. It won honestly by flying great and being flow with great skill. This was against a field of scratch built or short kits that where sporting the new short span wings. More on the short span wings later.

The airframe are rather restricted by the rules. This was done to allow kit manufactures the ability to offer us racers ARF race planes. It was hoped that a perspective new racer would be drawn the ARF concept. (Not need to learn any building secrets to have a competitive ship.) The rules are also trying to nullify the advantage those of us that know how to build fast models. Composite uses limited, trailing edges need to be thick, airfoils are thick all this to keep the speed down and give the novice racer a chance at having fun the first time out on the race course.

Along with a restrictive set of rules all designs need to be approved by the NMPRA before being legal. This is to control those of us that might try to make a racer that is compliant to the letter of the rules but not in the spirit of the rules. The NMPRA board gets a lot of grief for this but I have to thank them for taking on this thankless task.

There are three features that make the GP Proud Bird so good. First and foremost is the very wide landing gear track. The Proud Bird is the best ARF coming off the starting line allow the Proud Bird to be the first to pylon one. The second feature is that the Proud Bird has a swept leading edge. This appears to help keep the yaw while making the pylon turn under control. That is the Proud Bird is less likely to tuck into the ground coming around the pylons. And the third is that the Proud Bird uses torque rods to actuate the ailerons, cutting down on linkage induced drag. The only real down side to the Proud Bird is that she has a lot of ABS plastic parts that need to cut out and fitted. Of the EF-1 ARFs, the Proud Bird seems to take about twice as long to assemble compared to the E-Flite offerings. A feature I like is that the Proud Bird is offered in white allowing the racer to customize the graphics making it easier to differentiate your racer from the others while flying.
http://www.greatplanes.com/airplanes/gpma1260.html

The next best ARF is the E-Flite Shoestring. Its best feature is that it is a mid/ shoulder wing model. Shoulder wing models seem to roll and track better through the whole race coarse. I like shoulder wing racers. The only real weak point for the Shoestring is the narrow landing gear track! Along the same lines the aluminum sheet landing gear is far too easily bent. The narrow track makes it difficult to get a straight launch to pylon one. Also the fact that the landing gear is so easily bent, means that one doesn’t know which way the model will pull on the take off roll as the wheels will be pointed in a different direction after each landing. Aerodynamically the Shoestring is at a disadvantage as she has aileron linkage drag from the outboard wing mounted servos. I like the fact that the Shoestring comes set up with dual aileron servos as it makes setting up the ailerons for spoileron landings much easier than on the Proud Bird. Setting up the ailerons as spoileron helps make the landing so much easier resulting in less damage as one is less likely to over run the landing strip. Another thing that I don’t understand is why E-Flite has put a 22 gram pilots head in the model. This is just dead weight. I remove it! I like the Shoestring enough to own 3 of them.
http://www.e-fliterc.com/Products/Default.aspx?ProdID=EFL4205

World Models has two fine ARFs, the Scarlet Screamer and Outrageous these appear to be in the same league as the E-flite Shoestring being great shoulder mounted wing racers. But they too have too narrow a landing gear tack. I’ve never built one but having raced against them they have nothing that they need to apologize for.
https://www.theworldmodels.com/store/?action=productDetail&id=10115

As it looks like the Achilles heal for most of these ARF models is the landing gear. I do replace them on my racers. For those that like aluminum plate gear the Great Planes 09 landing gear seems to work out much better than the gear supplied by E-Flite. You will need to reshape it a bit in a vice and maybe trim it a bit with a hack saw. But the end result is a gear that is much stiffer and a bit wider than the one supplied by E-Flite.
http://www3.towerhobbies.com/cgi-bin/wti0001p?&W=002736957&I=LXJ920&P=K

Now I like glass gear, they are stronger more flexible and predictable. That is the wheels point the same way after each landing or the gear is broken. I make mine by cutting out a crude form in some scrap lumber. Then I layup a lot of fiberglass cording until I have a layup that is about 3mm to 4mm thick. I use a very slow hard epoxy in my lay up. Once the epoxy is cured I do a post cure heat treatment at 75°C. I then pop off the formed plate and sand away anything that doesn’t look like a landing gear. You might have noticed that I make my gear squattier (lower the nose). This helps reduce the prop effect (swing to the left) on take off.

All the best,
Konrad
Attached files Thumbnail(s):
Re: EF-1 Speed Secrets
Re: EF-1 Speed Secrets
Re: EF-1 Speed Secrets
Logged

Cut it twice and it's still too short!
Konrad
Platinum Member
******

Kudos: 29
Offline Offline

United States United States

Posts: 1,662


Topic starter
Measure twice cut once



Ignore
« Reply #7 on: July 01, 2017, 08:50:02 PM »

The short wing EF-1 has been a concern for some, usually those that don’t know what EF-1 is all about. Around 2014 the NMPRA dropped the minimum span rule. This was done to open up the variety of models being flown in EF-1. Mainly to allow the elliptical winged Sport F1 being flown at Reno to be eligible for EF-1. As it is understood that high aspect ratio wings will suffer from induced drag at a lower rate than a short aspect ratio wing. To help the short wing EF-1 to be competitive the NMPRA kept the airfoil thickness requirements the same for both the longed 52” max wing and any approved short wing design. It was expected that while the short span model would have less frontal area this would be balanced out in that the short winged model would suffer more from the induced drag, even more so in the turn. Time has show that the NMPRA board was correct. There is NO advantage in having ether a long winged or short winged model.  Three years after the change and it was a long winged model that won the 2016 NATS.

Where some folks get mixed up is that when the rules changed it was the fast, dedicated racing guys that went through the effort to make short winged models. So it appeared that the short wings were faster because the fast guys were flying them. The short wing aircraft is not inherently faster. They are fast because fast pilots are flying them.  So as an entry level pilot please have no concerns that the big box distributor's ARFs are at a disadvantage against the short wing elliptical racers. It just isn’t so, the record books show that the long winged model is very competitive.

I’m showing what I think is the best EF-1 out there. That is D. Gall’s Loki. The orange one is V1 with the long 52” wing. The much better looking, as it is closer to scale, white one is the short winged V2 Loki. Both fly great! If the white is faster it is because it is lighter, it has a true pressure cowl for cooling and the designer has learned a thing or two about streamlining. It pains me to say this but if I was flying the new white racer against Gall, flying his old long winged  Loki I would still get lapped twice in any given heat. The short wings are fast because the fast guys are flying them. I asked Gall why the change to the short wing. He said he liked that he could model a racer closer to a scale outline. He though the long winged Loki just looked wrong. I disagree but that’s for another thread.

As the pictures of the V2 Loki are a bit washed out I thought I'd post a better one. As you can see the V2 can collect some hardware at the NATS. This was despite the fact that Gall had some issues with energy management. He sport flies at 6K feet (Denver area) and coming down the sea level requires some adjustment in ones flying.
Attached files Thumbnail(s):
Re: EF-1 Speed Secrets
Re: EF-1 Speed Secrets
Re: EF-1 Speed Secrets
« Last Edit: July 01, 2017, 09:27:34 PM by Konrad » Logged

Cut it twice and it's still too short!
Konrad
Platinum Member
******

Kudos: 29
Offline Offline

United States United States

Posts: 1,662


Topic starter
Measure twice cut once



Ignore
« Reply #8 on: July 02, 2017, 05:55:16 PM »

Now we come to the equipment list. Again folks that haven’t flown the class try to find an advantage here. Let me state up front that if you choose any combination of NMPRA approved and recommended components you will NOT be equipment limited! The NMPRA only regulates 5 things for the  EF-1 Pylon class of racer. The airframe we have discussed that. The motors have to be approved motors. The prop, only the APC 8x8 is approved. And the batteries can to be any 4 cell 4.2v cell chemistry (16.9V max) lipo that cannot weigh no more that 335 gram with connector. The esc has to be 60 amps or larger and have a street price of less than $150 usd.
Here is the list of NMPRA approved and recommended components:
http://www.nmpra.net/EF1Approved.htm

Again to save one from having to suffer reading the whole post I’ll summarize, this is the combination most of the fast guys were using as of last year (2016 NATS).

Motor: Rimfire motor
ESC: Yep 80 amp
Batts: Venom 3600 30c

The motor all others are measured against is the E-Flite Power 25 1250KV. That is because it was the first good EF-1 motor easily available. It is still a very good competitive motor. But it is a bit heavy. Unless your plane needs the nose weight it might not be the very best for your application. (I still race one, but more to the point I’m constantly beat by this motor)!
http://www.e-fliterc.com/Products/Default.aspx?ProdID=EFLM4025B

The next motor that is very tempting to the novice racer is the Hobby King NTM-EF-1-1300KV V2. What draws folks to this motor is that it is so cheap and it has the highest KV of any of the approved motors. Unfortunately it is also the least efficient of all the approved motors. That means that it has the highest RPM sag as the motor sees a load. It also is the one with the odd dimension (form it has the largest diameter of the currently available EF-1 motors). And yes it has some quality issues. The cross mount still rubs against the lead wires so one will need to grind the needed clearance into the cross mount. Also the grub screw that holds the rotor to the motor shaft is easily stripped as the aluminum rotor end in made of a low grade aluminum. Since the motor is the least efficient it place the most stress on the batteries for any given power. This means it gets the hottest but since it has the largest air gap it responds the best to the pressure cowl. On the race course they pull as well as the other approved motor. (I own a few but will be replacing them as need be with the RimFire)
https://hobbyking.com/en_us/ntm-prop-drive-series-ef-1-pylon-racing-motor-1300kv-930w-v2.html

The newest EF-1 motor is the Cobra/Scorpion Motor. I have yet to race against or even see one of these. But based on the reputation of Scorpion it should be a very nice competitive motor. I would like it a lot more if it had a longer shaft to mount a collet spinner.
http://innov8tivedesigns.com/cobra-c-2826-ef1

A motor that you can sometimes find on the secondary market is the OS EF-1. This motor was not competitive ether on price or performance. The OS motors are just rebranded Chinese motors, likely SunRay ltd. So it was hard to justify the premium price of three time that of the competition. Also the OS motor had an effective fan. This fan robbed power from the shaft that would otherwise be used to turn the prop. OS has removed this motor as it was a disaster in the market place.

Great Planes now has two motors approved for EF-1. These are the Rimfire 35-45-1250KV EF-1 and Rimfire 35-45-1250KV EF-1-Collet. These motors are every bit as good as the E-Flite Power 25. The major advantage to the Rimfire is that the collet version winds up being 20 grams lighter when one adds the prop adaptor/spinner to the mix.
http://www3.towerhobbies.com/cgi-bin/wti0001p?&I=LXDGMS
http://www3.towerhobbies.com/cgi-bin/wti0001p?&I=LXFFMF&P=7

Now a real “Speed Secret” is the use of collet  spinners. They have the dual benefit of running truer than bolt on spinner, and as there are less joints and if made from metal, draw heat away from the center of the motor. This is true of inrunners that the center shaft will help draw heat from the magnets. With the outrunner motor this isn’t case but anything that might draw heat away is a benefit. I’ve had real good luck with the E-Flite collet spinners.

ESCs are a place where you might find some hidden power. As a general rule the higher the amp rating the lower the internal resistance for the ESC. This means that there is less of a voltage drop across the ESC and this  provides more voltage to the motor. More voltage means the motor can spin at a higher rpm. The trade off is that the higher amp rated ESC are usually heavier or cost a lot more. Most folks have settled on the 80 amp YEP/Rotor Star or the 75 amp Castle Creations Edge lite.

I use the YEP 80 as it only weighs 70 grams. But as of late I’ve been having a real issue with the switching BEC on the YEP ESCs. That is on plugging in the battery the BEC is allowing a voltage spike to pass through to the Rx and servos. This often burns up the microprocessors in the ESC, RX and dIgital servos. I now strongly recommend the Castle Creations 75 Edge lite. Even at double the cost of the YEP, the Edge lite 75 is a far better value than the YEP. And you can make up for the added 12 grams in weight by trimming the wires to length. With the weight savings from the newer generation batteries I might try some 100 amp ESC.
http://www.castlecreations.com/en/phoenix-edge-lite/phoenix-edge-lite-75-esc-010-0112-00
https://hobbyking.com/en_us/hobbyking-yep-80a-2-6s-sbec-brushless-speed-controller.html
https://hobbyking.com/en_us/rotorstar-80a-2-6s-sbec-brushless-speed-controller.html

Batteries, this is where all the magic is! On the subject of Lipo batteries there is an ugly secret that few talk about. That is, if the battery is more than 2 years old it isn’t suitable for powering our model aircraft. This is because the electrolyte in the lipo cell degrades the anode and cathode plates just sitting around. We try to slow or more to the point balance this deterioration by storing our batteries in a 1/2 charged state (3.85V). This is done to try to extend the life of our batteries. Unfortunately this fact of life also applies to the batteries used in EF-1 Pylon. So the fact that the battery technology is changing at a very fast rate really shouldn’t be of too much concern as you would have had to rotate out your batteries every 2 years anyway to be competitive.

At the 2016 NATS I recall these where the batteries of choice:
https://www.venompower.com/products/e-flite-carbon-z-scimitar-30c-4s-3600mah-14-8v-lipo-battery-by-venom

Graphene has entering the development of Lipos so this is a constantly changing area. It is looking like these are the go to batteries Dinogy 2600 mAh 70C. These have enough capacity and at 291 grams are a huge weight saving over last years 3300mAh battery of 335 grams
https://www.dinogylipos.com/collections/4s-lipos/products/2600mah-4s-70c

So for those that read this post the take away is this:
http://www.castlecreations.com/en/phoenix-edge-lite/phoenix-edge-lite-75-esc-010-0112-00
https://www.dinogylipos.com/collections/4s-lipos/products/2600mah-4s-70c
http://www3.towerhobbies.com/cgi-bin/wti0001p?&I=LXFFMF&P=7
https://www.horizonhobby.com/product/airplanes/airplane-accessories/spinners/175-aluminum-spinner-with-4mm---5mm-collets-eflsp175

I hope to see you on the winner’s podium!

All the best,
Konrad
« Last Edit: July 02, 2017, 06:10:10 PM by Konrad » Logged

Cut it twice and it's still too short!
Konrad
Platinum Member
******

Kudos: 29
Offline Offline

United States United States

Posts: 1,662


Topic starter
Measure twice cut once



Ignore
« Reply #9 on: July 08, 2017, 02:08:42 PM »

Servos are a place where some can find performance. I find that 13 to 19 gram servos are adequate for the low speeds that these EF-1 are flying. As  general rule I like my racing servos to have motors that are wound on a core. I’ve found that coreless motors don’t hold up as well to vibration.  With the wooden airframe structure, prop induced vibration can be an issue. This is why we need to balance all racing props. They may not contribute to speed but a balance prop extends the life of the radio components.

I try to use servos that have a torque rating of 3.0 Kg.cm or greater. The one exception is if using dual aileron servo set ups. I’m comfortable with aileron servo for the dual aileron servos installations being down in the 2 Kg.cm range.

Since I generally don’t want too fast a responding set up, servo speed is not much of a concern for me, as long as the servo has around .18 sec/60° or better response time. The one exception is the elevator servo I like the elevator servo to have a speed of 0.0x sec/60°

I also don’t care for metal geared servos as they have historically developed gear slop noticeable faster than the nylon geared servo. The one exception is for the rudder as this servo often does see a shock load as the model swing left and right on take off or is suddenly loaded on landing.

While these servos don’t meet all my racing requirements I’ve been using them to good effect in my EF-1 racers.

http://alofthobbies.com/daviga-ds213-servo-3-0kg-41-7-oz-in-07-sec.html

http://alofthobbies.com/emax-es09md-servo-2-6kg-36-oz-in-08-sec-14-8-grams.html

http://alofthobbies.com/emax-es3352-thin-digital-servo-2-8kg-39-oz-in-10-sec-12-4-grams.html
Logged

Cut it twice and it's still too short!
Pages: [1]   Go Up
Print
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Valid XHTML 1.0! Valid CSS!