Search

Next Generation Lightweight Cooling Fan - Part 2

In "Part 1", I described a couple of customer issues with the first generation lightweight cooling fan that caused me to rethink everything, and attempt to come up with a "Next Generation" lightweight cooling fan that I would have full confidence would always outperform the OEM fan under all conditions. In this "Part 2", I want to talk about some of the details of what I have done, and the process I followed to get to where I am, and in "Part 3" I'll talk about the results I achieved with the "Next Generation" fan.


Going back to the squirrel cage design


So, I went back to the squirrel cage centrifugal fan design, just like OEM, and as I described in "Part 1", this is to prevent fan performance from suffering too much to provide better cooling if there is a situation where the static pressure the fan has to overcome is higher than it would be normally for some reason. So, in going back to this design, that meant looking long and hard at all the components, since there just isn't enough room to just make a larger fan. With a squirrel cage centrifugal fan, the easiest way to get more air, is just have a taller blade. That is what VW did between the early engines, and the '71 and later engine and doghouse fan shroud. They went from a 28 mm blade height to a 34 mm blade height. Everything else is identical, and I can confirm this to be true as I cut apart a fan, and took many measurements between them. These blade measurements are just the blade, not counting the upper and lower sheet metal pieces, so don't be confused if you have seen different measurements from other sources. In going back to this design, and with the limitations of the stock components, that meant I had to take on not just designing a fan, but designing other components to make a larger, and more efficient fan fit in the same space.


The other components


In looking how everything fit together, the first thing I realized is that the backing plates take up quite a bit of room. They are designed the way they are to allow the fan to pull air through the generator or alternator, to help keep it cool and not overheat. So, anything new has to allow the same air flow through the generator/alternator while simultaneously allowing the same airflow through to cool it. The inner backing plate is the key piece, as it is takes up the majority of the room in the fan shroud that could be available for a fan. So, through some iterations, I designed a new backing plate that frees up around 7 mm of extra space while still allowing the same amount of airflow to be pulled through the generator/alternator to keep it cooled properly. In fact, there is more area for the air to flow though than on the stock backing plate, so there will not be an issue with an overheated generators or alternators. The next component is the mounting system.


As many of you know, with the first generation lightweight cooling fan I had to introduce a new mounting system because the potential of the fan cracking and failing. Since I had already done that, and I really couldn't change everything on the first generation, because I had already invested over $14,000 on a mold so I was somewhat restricted in my approach. Starting from scratch, with a new design, meant I could rethink that completely. With the original design, the fan mounts on a hub that has two flat sides and two radius or curved sides (I'm sure there is a technical name for this, but I don't know what it is), and that works great for stock engines that don't turn more than 4,500 RPM. With higher performing engines, it's a problem, and I have personally seen quite a few fans where that mount area of the fan gets rounded out, and the fan is no longer able to be driven properly, causing the engine to overheat. In thinking about this, all the force is essentially on four corners, which is why that area can get rounded out. It made me think of the way the flywheels are attached to the crankshaft, and how generally flywheels are attached in most higher performance engines. For the Type 1 engine, VW used four dowels and a single bolt, that we all call a gland nut. In the early days of building high performance Type 1 engines, this quickly proved inadequate, and then through experimentation they added four additional dowell pins and increased the torque a bit on the gland nut and also sometimes use a larger washer or nut to spread the load over a larger area. That proved adequate for most engines, and is the common thing we see today. If you look at later generation engines, VW, and virtually everyone else in the auto industry went to multiple bolts to attach flywheels. This spreads the pressure more evenly and distributes the loads across more points, making that the best way to attach a heavy and large diameter flywheel to an engine turning very high RPM's. With that in mind, I created a new hub, that is very low profile (more on that in a minute), and has six threaded holes for six bolts. This gives me a flywheel style mount, where the fan attaches to the hub with six bolts that spread the loads across more points and provides even pressure. Now, this does not hold the fan on to the generator/alternator, but just holds the fan to the hub, and the hub slides on the generator/alternator shaft. You still have to have a nut on top of that, and originally I designed a custom hex nut for that purpose, but that proved way too expensive. When you make any part that requires really tight tolerances that costs go way up, and hex nuts do require tight tolerances. So, instead I took the existing flanged nuts that I ship with the first generation lightweight cooling fan, and designed a spacer that could except a spring washer and the flanged nut. That provided some extra benefits, besides being much cheaper to manufacture and purchase.


I am making the spacer out of aluminum, and that means as it heats up it will expand considerably more than the steel hub and the generator/alternator shaft, making the entire assembly actually get tighter as you reach operating temperatures. That in combination with the fact the spacer takes a spring washer (like the OEM custom spring washer and huge nut), means that the need to use Loctite is removed, and the overall torque needed to keep the fan in place is lowered as well. So, the new mounting system is mechanically better in every way, and it is cheaper to manufacture than my mount on the first generation fan, even though it's composed of more pieces. Now, I mentioned that the mount was very low profile.


When I first designed the new six bolt hub, I used the same height dimensions as the hub that I shipped for the first generation lightweight cooling fan. I designed the fan in such a way that the center sitting on the hub to the edges would just curve down taking up the extra space that my new backing plate afforded, but that caused some interesting fan performance issues. When I started testing the fans on my test platform (which you can see detailed in my YouTube video here: https://youtu.be/UPfRVF--b40) I was getting poor results. In fact, my airflow was some 8% lower than OEM, even with the larger blade. It goes to show you, that you can't just create a fan with a larger blade and get better performance, there is more to it than that. Part of what was happening, along with a lot of other small details, is that deep curve was causing turbulent airlfow in the center of the fan as it tries to enter the blades, causing lost performance. So, making the hub as low profile as possible had the benefit of making it cheaper to manufacture (a lot less material), and the big benefit of improving the airflow performance of the fan. All the parts have to work together as a cohesive solution to get the best results.


What's coming in the kit?


The original lightweight cooling fan came with the fan, of course, the custom mounting hub, a stainless steel flat washer and a flanged nut, and some optional spacers that were laser cut from spring steel, in case you have rubbing issues with the backing plate. As it turns out, some generators or alternators have very different size aluminum spacers that the mount hub sits on, and they vary enough in size that I needed to provide these spacers. Most people didn't need them, from what I have been told, but some did. With that issue in mind, I also created my own aluminum spacer to set on the back of the generator/alternator, so I could control the tolerances, and know that you would get the right fit regardless of which generator or alternator (more common where the aftemarket alternators) you have.


With the "Next Generation" lightweight cooling fan you'll get the generator/alternator spacer made form 6061-T6 aluminum, the new inner backing plate, you'll get the six bolt mounting hub with the socket head bolts (6 mm x 1 mm), you'll get the fan, of course, and you'll get an aluminum spacer that sits on top of the fan and goes over the shaft of the generator/alternator along with a stainless steel spring washer and the same flanged nut from before. No need for any spacers to prevent rubbing, as I am controlling all the tolerances involved, so you won't have a "tolerance stacking" problem where things just don't fit because everything is either at the low end of the tolerances or the high end, and things just don't fit anymore!


Finally, there will be an optional piece of the kit, which is the velocity ring. I have comprehensive data on the velocity ring performance, and you get the very best performance out of the "Next Generation" lightweight cooling fan with a velocity ring in place. With that, let's talk about what everyone is really interested in, which is the fan!


The fan!


Let's get to the real meat here, the "Next Generation" fan. In talking about the other components that will come with the kit, the backing plate being much thinner is what allowed me to create a larger squirrel cage centrifugal fan, but as I mentioned before, just having a larger blade does not equal higher performance in all cases. Just like with the hub design, you can make a larger blade fan that under performs a smaller blade fan, and there are many ways to do that, as I discovered for myself.


So, the new Doghouse fan shroud fan is larger, and it's blade is 41.25 mm in height, compared to the 34 mm blade height of the original. Even with that increase in blade height, my initial designs performed worse than the OEM fan. So, I iterated over a number of different elements to get the most out of the fan, and eventually got a fan that outperforms the OEM fan by a substantial amount. Those elements were as follows:


  • Blade curve on the X,Y plain

  • Blade curve on the Z plain

  • Blade angle on the X, Y plain

  • Radius of the base of the fan from the mount hub to the start of the blades

  • Radius of the inlet of the fan to the top of the blade

  • The way the radius of the inlet of the fan intersects with the top of the blade

  • The design of the leading and trailing edges of the blades

  • The design of the leading edge of the fan inlet

  • The design of the inlet of the fan in relation to the inlet of the fan shroud


All of these design elements made a significant contribution to the overall performance of the fan, and VW engineers did a great job with the original OEM fan, that is for sure. All of the above elements are different on the "Next Generation" lightweight cooling fan compared to the OEM fan. It took me a lot of iterations to optimize all of these elements, but what we ended up with was a fan that performs really well, but with one thing that surprised me quite a bit. The surprise was the performance with a velocity ring. When you add a velocity ring to a Doghouse fan shroud with the OEM fan, you get a 5.25% increase in airflow. A very nice increase indeed. When you add a velocity ring to the "Next Generation" lightweight cooling fan, and in fact the same velocity ring, you get a little over 9% increase in airflow (9.04% to be exact)! That's over a 3.25% additional increase over the already increased airflow of the new fan! All the small optimizations over the OEM fan, get magnified when you reduce the inlet pressure, and the "Next Generation" lightweight cooling fan just comes into its own. It is a beautiful thing to behold as an engineer. When I first starting recording the data I couldn't believe my eyes, but it was completely reproducible across tests, so the measurements were not lying ;-) In any case, I think everyone will be really pleased with the "Next Generation" lightweight cooling fan, and I'll be confident that it always outperforms the OEM fan in all conditions, regardless of what modifications have been done to their vehicle, what year their vehicle is.

60 views0 comments

Recent Posts

See All

Why a "Next Generation" Lightweight Cooling Fan? My first generation lightweight cooling fan was a straight radial blade centrifugal design. The reason I chose that design was because the internal st