Individual Throttle Bodies and Improved Airflow

[Dessertrunner]

Originally Posted by longassname

It amazes me I still don't own one of these. I was eyeing the SPI version back when it took me several tries over several days to get the spindle on my machine truly square. I'm going to redo some stuff on my machine and think I'll get one this go around.

Have you used a mist system that was effective at clearing chips out of your parts?

Haven't tried misting, I figure it would not work on Alum. Even with flood coolant when cutting with 4 flute 20mm dia at 6mm deep it has trouble stopping the tip sticking to the alum.
Maybe I don't understand.


On the throttles I ordered a reamer at 35mm and will open the back side up, go to using 2 side plate instead of 1. In the same way Adam did, also mount bearings in it carry the roller. May stick with the brass a bit longer. Its easy to cut so when I have sorted the current problems out I will see if its possiable to lighten them.

Spoke to Mark (air bench guy) and we will hopfully put one on the air bench next week to compare the flow curve to a normal throttle body such as a SVX. The reason is that a number of people have pointed out that the roller throttle are terriable at low speed/part throttle, very hard to handle. If you remember I asked the question should they open in the same direction or opposites well there is a 3rd option and that is should one open fully before the other.
As they say just because they have done it one way for years doesn't make it right, we need to question it, also we need data to know what we are talking about.

Have a great day.
Tony

[Dessertrunner]

okay the brass worked out a lot better then I expected I have gone to bearings on the end of shaft. When I decide which way the barrels turn I will mill the back to lighten them up. Also want to wait for the reamer before I make the rest of the rollers and the bearing carriers. Hopfully the ram pipes will arive this week coming as well.

Here is a photo of the new design with the milled brass rollers after reaming the bodies I will check the amount of clearance when they get hot.
Tony

[longassname]

I have a graymills mist system on my machine now and haven't had any problems with micro-welding with a big end mill like that. I did have bad welding problems with very small 4 flute end mills cutting a negative relief on a very large ship's number plate where I was really really trying to keep the speed up while minimizing cutting forces so as not to snap the small diameter tools and still finish before growing old. It's not every day I'm trying to cover a large area with tiny tools though. My real complaint with the sytem is it doesn't clear chips well so I can get a lot of re-cutting.

The o-ring on the valve stems on the manifold of the graymills system went bad and instead of replacing them I want to try a new system. My table has plumbing for flood drains but I'm giving mist another shot. I just ordered a kool mist 102ff18 which has two flex nozzles and spec's a max of 125 psi. The graymills spec'd 60 to 70 psi. I'm hoping twice the pressure and twice the nozzles will be enough to blow the chips away to my satisfaction.

How are you generating your g-code?

Originally Posted by Dessertrunner

Haven't tried misting, I figure it would not work on Alum. Even with flood coolant when cutting with 4 flute 20mm dia at 6mm deep it has trouble stopping the tip sticking to the alum.
Maybe I don't understand.

[Dessertrunner]

I am interested to hear how it goes for you with the misting, getting rid of the chips is my biggest problem on the job, you have zero tolerance to chip drag on the roller points.

I am using a program called CamBam
http://www.cambam.info/
its cheap and will do 3d as well. I then run the code in Cutviewer to check before running on mill.

By the way I am happy to send anyone who want the GCode or design for the throttle, would be great if some of you guys out there jumped in on the development as it would be faster.
The final design will if want be able to run on the standard ECU or if desired one of you cards Micheal.
Tony

[longassname]

Does cambam remember what material has been removed by previous tool paths?

I have cut2d from vetric which is great except it does not remember what material is removed by previous tool paths. So for example on that job I was talking about where I was cutting the material out around letters to leave raised letters inside of a recessed plane on a large plate I couldn't first use a large tool to remove most of the material and then generate another toolpath to use a smaller tool to get the detail around and inside of the letters/numbers. Vcarve pro does do that but it's another $480.

[Dessertrunner]

You can down load a trail version, my guess is you can get around the problem by using pocket & profile using different clearance.

[Dessertrunner]

You can down load a trail version, my guess is you can get around the problem by using pocket & profile using different clearance.

Just realized you can use boundry box to also get around it.
Tony

[longassname]

Ya, I've done the work around thing just to get the job done but it's not the same thing. It works better than just trying to use too small a tool for the whole thing but it isn't as efficient as if the cam processor took previous toolpaths into account when generating the next. Taking that number plate as an example again the workaround method was a pocket toolpath with a 3/8 endmill, then a profile toolpath with 3/16 endmill, then a profile toolpath with a 1/8 endmill. If I had vcarve pro I could have generated the 3/8 pocket toolpath and then a 1/8 pocket toolpath and it would remember all the material removed by the 3/8 toolpath so the 1/8 endmill toolpath would just take off the bits left over. That would be half the runtime.

Originally Posted by Dessertrunner

You can down load a trail version, my guess is you can get around the problem by using pocket & profile using different clearance.

Just realized you can use boundry box to also get around it.
Tony

[longassname]

Here's a picture of the plate to help you visuallize what I'm saying. Having to run two complete profile toolpaths on this was a lot of wasted run time. Most of the time it wasn't cutting anything but then in some of the corners there were still full profile cuts so I had to make shallow cuts so as not to snap the small diameter tools. That's like 6 passes all the way around with those little end mills just to get at the nooks and crannies to detail the raised letters.

All that wasted run time peeves me.

http://www.ecutune.com/mill/DSC_7940.JPG

[longassname]

I was just looking at the photo of the single itb you have going there and have an idea you may want to pursue. You could skip making side plates, using bearings, machining the ends of each roller, and the gears on the ends of the rollers for each ITB. I would just make plain bores in the ITBs and instead of individual "rollers" with gears between them/the neighboring ITB use a plastic rod all the way through all 3. All you have to do is cut grooves in the rod for sealing rings. That's much less machine work, much less hardware, and a cleaner, lighter end product. You should only have to make an end plate and mount gears at the end where you have to do something about the linkage anyway.

[Dessertrunner]

I wasn't going to use gears that Adams design, your idea to connect them all together is a good suggestion. My question is what type of plastic can stand the heat?
In the last 24hours I have come up with a redesign on the throttle bodies that will enable the injectors to be fitted. In addiation I can drop the height by 10mm which will help with the air box.

I will post the new design in the next day or so.
Have a great easter all.
Tony

[longassname]

maybe garolite g-11?

[oab_au]

Originally Posted by Dessertrunner

Spoke to Mark (air bench guy) and we will hopfully put one on the air bench next week to compare the flow curve to a normal throttle body such as a SVX. The reason is that a number of people have pointed out that the roller throttle are terriable at low speed/part throttle, very hard to handle. If you remember I asked the question should they open in the same direction or opposites well there is a 3rd option and that is should one open fully before the other.
As they say just because they have done it one way for years doesn't make it right, we need to question it, also we need data to know what we are talking about.

Have a great day.
Tony

Tony the I think the main function that you need to achieve is to make the throttle opening, 'air passage', progressive as the rollers open, just as the butterfly does. A small increase at first, increasing as the throttle is opened. Makes for a better, smoother response.

Harvey.

[Dessertrunner]

Hi Harvey just the bloke I wanted to speak to,
All you old carbarutor guys can tell me why cars had 2 and 4 barrels.

I agree with your point Harvey and think the problem with roller throttles is they are either on or not on no middle or controled slow opening. So to my question, I want to open one roller first then when its fully open open the other one simlar to what will happen in a multi barrel carb. Also to explain I will open the left one counter clockwise and the right clockwise on the second stage. As the left roller opens it force the air to travel a longer distance as its going over the top of the right throttle then turn and go into the intake. Will post some roller images later of what I mean.

I have solve the injecter problem and cut 30mm off the current injector rail height so it should be easier to fit the airbox longer ram pipes. will post later.

So Harvey and everone else what do you think?
Tony

[oab_au]

Compound throat carbs were used to ensure a good depression at the jet for fuel atomization. You don't have that problem with injection, as long as the low throttle opening air stream, is aimed at the injector.

The flat slide throttles had a problem with a non progressive opening, really only good for racing, hard to drive on the road.

Don't know which way to open the rollers.

Harvey.