Jet engine Shaft Balancing
Hi again, as you may know i'm building a jet engine but not just any engine. I'm starting from scratch and the shaft is a little over a foot long with 2 special bearings in the middle. I am making parts out of aluminum then i plan to cast wax or plastic w/e works molds of the parts and then pour some molds with alloy metal or w/e i find at the store that meets my req. But after that i'm going to need to do final tolorences and then balance the shaft. I plan for a 120,000rpm. I don't have much knowledge with balancing things. What have you guys done to balance your engines and what kind of machines or methods have worked best for you. Do you know of any devices you can make or buy that will be reliable and accurate? It doesn't have to be the most accurate blancing job in the world just what ever will balance it enough that it will scream along nicely and won't blow into peices. Or what ever will balance it that it will work?
Wow a foot+ long shaft, now that's got to be a big engine.
Is it a two piece hollow shaft?
Just leave all the balancing for the last steps in making your turbine. Get all individual parts balanced reasonably close then balance everything as a complete unit, turbine wheel, compressor, shaft, nuts, etc.
I got all my work done at a turbo place as they had all the expensive balancing gear. final balancing was performed with the turbine assembled with diffuser and rear cone removed.
Go and have a chat with a few of your local turbo builders and tell them what your doing and ask how you should go about it.
These guys balance parts that are turning in excess of 200,000 rpm daily so most of them really know there stuff. My engine was balanced for $50US.
Not really worth stuffing about.
But you might find this interesting if you are really keen.
The Compressor is about 7 inches long, the back turbine is about 4 inches long and the middle shaft that connects the 2 peices is about 7 inches long. The bearings and bearing housing take up most of the middle and the rest is taken up extending past the flame tube to the back 4 inch peice. In total theres 2 peices, the compressor and turbine are solid peices with about 4 of half inch shaft running threw the middlei connect them with a housing and run 2 pins threw it to make them 1 solid spinning shaft. I could connect them and balance the whole 1 peice shaft which i would like to do cause then my final design is all 100 percent balanced or i could balance the 2 peices seperatly. I'm guna call around and try to find a shop to do it cuase its that or buy a machine which would be expensive. I'll end up asking them if i can find a place where i'm at. Its a 2 peice solid shaft. I don't have the luxury of building fancy peices. I have a small combo machine, the grizzly g0516 combo lathe milling machine. No cnc But if i could find a relitivily cheap balancing machine that would be great cause then if i build more i don't have to rely on other means for my production line?? My compressor is going to have about 96 blades total. 16 blades all the way around and 6 rows of them, is that too much? they will prolly be at a 45 degree angle cause thats the halfway point both ways, might be the most efficient angle. I don't think 96 blades is too much the whole engine relys on how big the compressor is. It will be intresting to see if this engine is a sucsess. I'm still machining the flame tube right now, then i have a couple more peices that are pretty easy, a couple days worth of machining, then casting time, then final touches, then rpm circuit which no1 is able to help me with, then startup!!!! I've got a nice gokart frame i'm testing the engine on hahahaha
The cheapest balancer made is a static prop balancer for hobby airplanes - hardly what anyone familiar with high speed balancing would consider even marginal for balancing a 120,000 rpm jet turbine. Check any of the model plane books and you should be able to find it.
After you spend that much time and money crafting up a compressor and assembling it, it might simply be smart to spend $50 to have a pro balance it - money well spent all things considered.
"Relatively cheap balancing machine" are oxymoronic terms when used in that order to describe the equipment that you're looking to duplicate/obtain "on the cheap".
BTW: the skills and math needed to calculate the forces involved in an unbalanced compressor turbine running at 120,000 RPM dwarf the calcs needed to do the fairly simple resistor and capacitor timing for the 2907 tach circuit you asked for help on/with.
Speaking of which, your comment about "no one being able to help me with" your tach circuit is TOTAL, ABSOLUTE AND UNMITIGATED B/S!!!!!
Pastera, others and myself gave plenty of help and/or direction on where to find circuits and how to get around some of the challenges of making that DIY circuit work in your multiple tach thread inquiries.
Pastera offered to check your calcs for a 2907 circuit if you didn't know enough about electronics to do so yourself. He even almost offered to cut you a PCB. All he asked you to do was do some prelim calcs and TRY to figure it out. Did you???
The impression given was that you apparently did not want to GOYA and do even the most rudimentary of calculations to even ATTEMPT to build a DIY tach circuit.
You can hardly claim a proven circuit and the IC that is based upon does not have "promise" if you won't take the time to GOYA and learn how to use it....
Under the circumstances, do not EVEN DARE to suggest that "no 1 is able to help me with" the tach circuit. Several folks tried and it was more like "I''m too lazy to build it myself because the math was too hard to figure out".
Yes, no one was willing to design, create and/or build and hand the tach to you - that is what people who make commercial tachometers get paid to do - and the purchase of a commercially made jet engine tach was clearly not something that you expressed an interest in obtaining.
Looks like sum1 is trying a little to hard to make himself feel good. This has nothing to do with a rpm circuit. i said no1 is able to help me cause i asked for a simply circuit and u went over board with all your 29017 ic's. Theres hardly any math and calculations invovled with balancing a shaft what are u talking about mr.i know everything. i know about balancing prop shafts theres no math idiot. with a prop its 2 blades, i'm taking about 96 blades around a shaft thats 3D. prop shaft is less then an inch if your using a big engine, this is over a foot long read the title. I was asking if any1 had a rpm circuit for me to build, ya thats what commercial ppl do but some people are nice and are happy to share knowledge u anger unhappy person. Its not b/s no1 was able to help me with my design i still don't have a circuit i'm still looking, no1 was able to help me with a final design i could just build and my rpm problem would be solved. Why are you still tlaking about the 2907 circuit i told you i wasn't intrested in that one anymore, i said i was going for optical because thats the best way to go. let me say that again but in another way incase u don't get it, Optical sensors work out best and thats what i'm intrested in... better?? Open your eyes and maybe u'd see not every1 is a electrical engineer and theres a huge number of ppl who don't have any idea when it comes to mhz, freq. , and other electronic related parts let alone the mathematics of all the parts working together, i didn't ask you to direct me to learn a whole **** load all i was asking was if any1 could share the knowledge of a simple rpm circuit , not for there rpm circuit design that they are manufaturing, loser. Anyways back to the topic of this tread!!!!!!!! Any1 else know of balancing techniques you've found worked or know of?? Any machines available for sale you know of? I've found 1 website that sells small parts balancing machine, it comes with a built in computer and software.They don't display the pricing so i don't know what its worth. BTW beaker thats a pretty impressive little homemade machine on that link, thanks for that
O and if you don't have anything nice to say don't say anything!! that goes to every1.... not just NC Cams
TWO engineers offered to help you integrate your optical sensor into a DIY tach - one of them (Aaron) even offered to check your math.
Did you take up Aaron's offer ? NO.
Oh and with regard to "optical sensors" being better at sensing RPM, neither Aaron nor I ever contended that optics DIDN'T sense better or worse than anything else.
I merely pointed out the technical FACT in non-electrospeak language that the frequency output by the sensor has to be connected to SOMETHING (ergo a tach circuit). The optic signal pulse train has to be changed into an electrical signal that is then read on a gage to indicate RPM.
Thems the facts Smurf - if you want to build a DIY tach, you simply HAVE to deal with frequency conversion or else go buy something.
It seems that all you wanted was someone to layout and give you the EXACT circuit to build with little to no effort on your part.
When that didn't happen, you then go on another thread and claim "...no1 is able to help me with..." which is a categorically FALSE statement and/or contention.
The respondants were ABLE to help but they were NOT going to do the project for you.
When you come on a message board and make false statements about the contributing members, you can at least expect some of the members to take offense and that is exactly what happened here.
Perhaps you might want to take some of your own advice, namely "...if you don't have anything nice to say don't say anything!!..." especially when you FALSELY contend that "no1 was able to help".
Oh, and balancing stuff to live at 120,000 rpm is not easy. I already have experience trying to DIY that sort of task when we played with R/C motor arms at 60K rpm and that was a headache - heck, when someone walked across the floor the "balance" was affected. The inbalance forces increase with the square of the speed and even the tiniest of imbalance becomes a massive force at 120,000 rpm. You need to be careful. We learned that lesson dodging 60k rpm R/C armatures that bounced out of our "balancer".
The point I"m really trying to make is this: Don't come asking for help on a M/B and when it is not given to you in the manner you want it, go elsewhere on the sit and bad mouth the percieved lack of support (Eg: "no1 was able to help").
Like you said, "if you don't have anything nice to say don't say anything!! "
smurf, i dont think nc cams is an idiot and or a loser
I am part of the Stealth twin turbo/3000gt vr4 comunity. You would not believe how many turbos fail even when they are balanced by "experts".
Make sure you find a good shop - do your research.
Beaker and Samco are right - what ever you do build, if it's going to run at 120k or even a third of that get it professionally balanced. Cheaper and safer in the long run. These speeds are not somethign to play with - where did 120k come from anyway? Its usually derived as a function of compressor design, efficiency and turbine section and application. Is this a thrust producing or shaft output producing engine? Design will vary accordingly
If you have 'real' high speed bearings for a turbomech application they should have a vibration spec in mm/sec or in/sec. That should give you a starting point. Plan to balance/weight match individual components, balance minor, major and full rotating assemblies. If your bearings are also inboard of the turbine sections then you've got to consider shaft flex too.
120k is a serious speed. Do I have it right, you're planning on casting the blades? FYI at 120k a typical section 12" compressor disk's centripital load could easily be greater than cast Aluminiums yield strength. Casting the shaft will be a balance problem. As cast the materials density will vary enough to make it unbalanceable to the degree required at your speeds.
If you're casting any of this in steel let us know how...
'prolly 45 degrees' for blade angle needs some research too. There's some fairly complex but readily available math that calculates the compressor and turbine sections blade angle of attack, shape, twist, losses and all that goes with it. AMSE had some good books on the subject, I found something on NASA while looking at stuff for the axial supercharger thread that had a free program to run the calc on axial sections too. Get googling..
Last edited by fyffe555; 11-03-2006 at 12:51 PM.
After going back and reading this I am not sure if I want to answer. To start with the shaft on mine is 5/8" (KJ66) rated for 125k. It is quite a bit shorter. Before you send out for balancing you can do a fair amount with what you have. Use a degreased set of brg's lubricated with kero or some other very light oil. Install the bearings and mount them. Using air to rotate the assembly and mark where it stops. Keep doing this and removing material till it does not stop in the same place. It should be random. It does not take much air and actually blowing on it should rotate it.
One safety comment is make sure the outer case is bullet proof or will contain the shrapnel if it lets go. Start slow and work up in rpm's. I think your estimate of 120k is a little high as the smaller ones are higher but the larger you go the slower they are. Lot of forces playing on your parts.
Ya i know 120,000 isn't somehting you should play with, my sizes of parts i gave you guys isn't exact i didn't have the plans infront of me, they were just a ruff estimate. 120 is a little high i don't know if it will acually go that fast but it doesn't hurt to plan ahead and be prepared for it to go that high, i don't want it going over 120, thats why i want a tach so it doens't go over that, 120 is extreme and anything over it is even more extreme. I relise that a long shaft will experience flex and that flex would be deadly at high speeds thats why i plan to get extremely strong metal, its hard to describe all the parts and there sizes to you guys and give you the right pictures in your heads, when its done i'll post pics and hopefully a video. Yes i'm casting the blades its going to be 2 solid peices spinning. Yes i know in casting the blades might be heavier on one side thats why i'm asking about balancing, i don't plan on spinning the engine untill i know its going to be safe. And yes the wall is going to be 1/4 thick and will be made out of strong strong metal incase something does go bad inside. A 45 degree angle to me is the halfway point of both ways, and might be the best efficient spot. Yes i could do math and calculations but i don't see 45 hurting anything, i've looked at lots of engines there blades vary from 10-45 from 45-80 estimate, i've seen blades in tons of arrangments so i don't think 45 will hurt. Yes i could do math but thats a engineers way, i'm just a home hobbiest building something cool doing it MY way... key word my way. Don't get me wrong i'm here asking for your opinions so i can have more angles to look at things. The more info the better. I plan to cast everything in alloy metal, or somthing that is as strong as specialized metal, i won't cheap out on metal, and the only aluminum parts is the bearing housing. Do you guys think i need to machine the bearing housing out of stronger metal then aluminum, i don't see why aluminum wouldn't work for the bearing housing, but then again its not much problem doing it out of steel, just aluminum is easier to machine what do you guys think? And NC Cams, Again i'll say it, i asked for a simply circuit, already designed, ready to go for me to use. I didn't take arrons help cause i don't ahve to cause i decided optical moron "Oh and with regard to "optical sensors" being better at sensing RPM, neither Aaron nor I ever contended that optics DIDN'T sense better or worse than anything else." I don't care what u think i have the final say and i said i like optical... get your facts straight before u go off on people, again get back to the topic of this threat don't waste my time with your little combacks and little efforts to make yourself feel better, if you don't ahve anything nice to say don't say anything, this thread is about balancing shafts, start your own thread called "i'm better then smurf and this is why" and go off as much as you want. No 1 was able to help me thats not false stop with your little bickering grow up, no1 was able to help me, my problemw as i didn't have a rpm circuit design to build, i still don't, you only tried to point me in a direction that was pointless and i'm not wasting my time on your stupid idea's. I simply said freq-to-volt isn't the way i'm going. I simply don't FEEL like doing it, cause i'm lazy and i don't care. Back to this threads topic please for the last time shut up with your stupid feelings.