Puppeteer

Author Topic: Design of steel frame mount and vibration isolation  (Read 80403 times)

snail

  • Full Member
  • ***
  • Posts: 243
    • View Profile
Re: Design of steel frame mount and vibration isolation
« Reply #30 on: October 03, 2006, 06:50:23 AM »
Guy,
     Piston complete 4.100 kg (Dursley 10/2, 1940)

     Rod complete 3.560 kg

I think you'll need to know the weight of the "reciprocating " and "rotating" parts of the rod. Haven't got that far yet. Likewise it should be possible to weigh the cast-in counterweight in the flywheels.It's on my "to do" list ;)

Cheers

Brian

Guy_Incognito

  • Sr. Member
  • ****
  • Posts: 255
  • Just a guy, incognito.
    • View Profile
Re: Design of steel frame mount and vibration isolation
« Reply #31 on: October 03, 2006, 08:11:40 AM »
Thanks Brian,
I'll have to presume a suitable fudge factor for the time being. Ummmm.... 70% towards the big end side?

I'll make it..... 5.1kg or thereabouts.
So that gives me (pauses to make a spreadsheet up) ..... about 0.7mm deflection at 650RPM. The force exerted on the ground through the mounts from your reciprocating mass every engine revolution is 0.75kg.

Just as a bit of an eye opener, I'll run this example with numbers for a typical rubber mount that compresses 5mm when you put 250kg on it. Four of them would hold an engine and generator. The k (see my original calc post on page 1) for those is worked out like so :

250kg gives you 5mm deflection, so 250*200 kg gives you 1 meter deflection , times 9.81 to get newtons gives you a k of 490500.
I'll pick a typical damping factor of 0.2 for internal friction in the mount. Changing nothing else in my equation except for that now gives me a jiggle of 1.4mm. Not much you say? The force exerted on the ground is now about 40kg every engine revolution.

I'll polish up my spreadsheet a bit over the next few days and I'll post it somewhere so people can play with some numbers and get an idea of what might happen with the mounts they've got.

Edit: I'll point out again that this is all theory, but allows me to get a ballpark figure. If I go build this as spec'd and I get 5 times the movement, I'll know there's something wrong and off I go trying to suss out what it is. If you don't do the calaculations, you'll never know what kind of results you should get and you're back to trial and error.
« Last Edit: October 03, 2006, 10:02:28 AM by Guy_Incognito »

hotater

  • Hero Member
  • *****
  • Posts: 1557
    • View Profile
Re: Design of steel frame mount and vibration isolation
« Reply #32 on: October 03, 2006, 03:14:26 PM »
Quote
If you don't do the calaculations, you'll never know what kind of results you should get and you're back to trial and error.

I'd rather do concrete than math.
7200 hrs on 6-1/5Kw, FuKing Listeroid,
Currently running PS-Kit 6-1/5Kw...and some MPs and Chanfas and diesel snowplows and trucks and stuff.

Guy_Incognito

  • Sr. Member
  • ****
  • Posts: 255
  • Just a guy, incognito.
    • View Profile
Re: Design of steel frame mount and vibration isolation
« Reply #33 on: October 03, 2006, 05:17:20 PM »
If you'd rather do concrete hotater, you're in the wrong thread  ;)

GIII

  • Jr. Member
  • **
  • Posts: 67
    • View Profile
Re: Design of steel frame mount and vibration isolation
« Reply #34 on: October 04, 2006, 03:05:20 AM »

xyzer

  • Hero Member
  • *****
  • Posts: 1058
    • View Profile
Re: Design of steel frame mount and vibration isolation
« Reply #35 on: October 04, 2006, 05:42:44 AM »
Guy.....actual weights from a Vidhata 6/1

Small end of Rod     1216   grams= 2.680#

Piston assembly      4025   grams= 8.873#
                  Total  5241   grams=11.554#
           *.5=50%   2620.5 grams= 5.777#

Big end of Rod       2742   grams= 6.045#
50 % of recip. mass  2620.5 grams= 5.777#
Vidhata 6/1 portable
Power Solutions portable 6/1
Z482 KUBOTA

Guy_Incognito

  • Sr. Member
  • ****
  • Posts: 255
  • Just a guy, incognito.
    • View Profile
Re: Design of steel frame mount and vibration isolation
« Reply #36 on: October 04, 2006, 07:11:08 AM »
GIII - They look alright, particularly the stops built into them, but if you were going to use them , you'd want to double check a few things.

The two very important things to remember with any sort of rubber/spring mount system are :

Resonance and engine speed.


I know, I'm starting to harp on about it but I can't stress this enough. As Guy_Fawkes has pointed out in the "how to kill yourself with a listeroid" thread, there are large amounts of force and energy lurking in your engine. 10 minutes with a pocket calculator will tell you whether it's reasonably safe to proceed with the mounts you plan to use.

Unless you have a very firm mount, your engine will pass through a resonance point as it spins up and down. If you can't get your mount to resonate above the maximum RPM of your engine (and I'd suggest it'd be a long way past your rated max RPM), you need to make sure it resonates at the lowest frequency possible. Two things help with that - 'bendy' mounts and lots of mass. You can substitute one for the other if need be, or use both.

I can spin my engine over by hand faster than the resonant point in the system that I'm going to use - it's point is calculated at about 70-80RPM. It's a low enough speed that the imbalance forces in your engine are not particularly large (due to v-squared). It's safe because the engine once started, will fire once or twice and it will be well above that speed then and off up to 650RPM.

Looking at the PDF brochure, those Lo-Rez mounts resonate at about 133 CPM (counts per minute - equivalent to engine RPM) , which is pretty good for a spring mount. Adding weight on top of that mount lowers the resonant frequency of the setup. The more weight, the lower the frequency (and the lower the up-and-down movement at normal speeds as well) Remember that there will be a smaller resonant point at twice the RPM noted due to firing pulses being at only half engine speed.

Any more than that - eg 200CPM , and I'd start to get concerned. Once you reach the area of 200-650RPM, the forces applied go up pretty fast. At 650RPM, you're  getting in the order of 150 kilograms of force pushing up and down on your engine and frame due to the piston mass alone. 10 times a second. This is why you need a solid frame.

A dangerous situation is having mounts that would resonate at 600-650RPM. You've got a large amount of force that will - if you have no bump stops or restrictions - exercise your mount and frame to destruction if it resonates at that speed. But there is a worse one - a system that resonates at 800RPM. You get an overspeed for some reason, things are already loaded just because of the rotational forces, and now you hit a resonant point.... that machine will be chasing you around the shop, if it doesn't fly apart when it drops off the mounts. For safety's sake, keep any resonance below 200RPM.  This is my concern with the try-it-and-see approach. Works fine at 650RPM..... at 1100RPM one day with the rack gets wedged at full fuel, it tears itself to bits.

If you can't figure out the natural resonance of the mount yourself, you need to ask someone. Ask the manufacturer. If they can tell you the deflection under load, you can work the resonance out. Ask me, if they don't have a clue - I may be able to help.

But having said all that GIII   :D - going off the criteria I've specified, those mounts look reasonably suitable, especially their bump-stop arrangement built in. Wonder how much they are?

Quote
Guy.....actual weights from a Vidhata 6/1

Thanks for those weights xyzer, I'll plug them into a spreadsheet I've got at work tonight. It spits out a graph of deflection Vs engine RPM which is very enlightening, so I'll see if I can post a shot of it tonight.

I'll double-check my spreadsheet over the next few days with the vibration engineer I've got on the case, and let you guys play with it then.
If you feed it a couple of things  - the mass of your setup, the spring rate of your mounts, a damping factor, it will show you a nice set of graphs from 10- 800rpm which have :

 - The amount of imbalance force applied to the frame and mounts.
 - The amount of movement (up and down) in mm because of that force. You can spot the resonant point pretty quick, it's the great big spike in your graph. My system will end up with about 1.5 to 2mm of jiggling from the looks of things. A bit more with torque pulses under load.
 - The amount of force transmitted to the ground, on the other side of the mount. Due to a curious quirk in the physics of it all, you can actually have a lot more force transmitted to the ground at certain frequencies near resonance than if it was just a solid mount.

This kind of info really helps in the design in a lot of ways.  Seeing when your engine is about to leap from the frame is one good reason of course.  ;D  But there are others reasons to do the maths.

For me, I'd want a frame that can not only carry the weight of the engine, but the weight of the engine + ten times the imbalance forces applied. So if I build a frame, set it up on a solid mount and apply (150 x 10) + 750 kilograms on it and it bends more than 'a bit' ... I'd be making something stronger. "A bit' , is something I'd like to define down the track a little more.  ;)

I know, it looks like overkill. But as someone famously said, you can never have too much overkill. I did question the frame design as suggested by mobile_bob originally, but looking at the forces applied, it's better to be on the safe side.

(And man! I need to lay off the coffee or something! Pretty wordy post.  :) )

GuyFawkes

  • Hero Member
  • *****
  • Posts: 1184
    • View Profile
    • stuff
Re: Design of steel frame mount and vibration isolation
« Reply #37 on: October 04, 2006, 03:17:18 PM »

Any more than that - eg 200CPM , and I'd start to get concerned. Once you reach the area of 200-650RPM, the forces applied go up pretty fast. At 650RPM, you're  getting in the order of 150 kilograms of force pushing up and down on your engine and frame due to the piston mass alone. 10 times a second. This is why you need a solid frame.



See, despite the fact you called me a luddite in the other thread, the penny is almost ready to drop....

7.5 approx kilos of piston and rod, just for the sake of argument and because it makes the numbers easy, lets make that 5 kilos of reciprocating mass, and 2.5 kilos of rotating mass.

Bore and stroke of a 6/1 is 4.5" by 5.5" in metric that's 114.3mm x 138.7mm

So your recoprocating mass of 5 kilos  travels up and down 114.3 mm 650 times per minute.

So your rotating mass of 2.5 kilos rotates around the crankshaft axis with a radius of 69.35 mm 650 times per minute.

This is basically a cylinder head off calculation, no compression, no ignition, no power stroke or torque, and these are not inconsiderable loads, in fact they can dwarf these loads, but these loads are simple to calculate, and lawd knoes we gotta keep it simple.

Don't forget doubling the RPM quadruples the load.

Let me see your calculations for the load exerted by the reciprocating mass using the numbers given above, and your calculations for the load exerted by the rotating mass using the numbers given above.

When you have done that we can move on to vectors and harmonics, one simple step at a time.

Remember, not just a number, I want to see workings and method.
--
Original Lister CS 6/1 Start-o-matic 2.5 Kw (radiator conversion)
3Kw 130 VDC Dynamo to be added. (compressor + hyd pump)
Original Lister D, megasquirt multifuel project, compressor and truck alternator.
Current status - project / standby, Fuel, good old pump diesel.

Guy_Incognito

  • Sr. Member
  • ****
  • Posts: 255
  • Just a guy, incognito.
    • View Profile
Re: Design of steel frame mount and vibration isolation
« Reply #38 on: October 04, 2006, 04:28:15 PM »
Oh,hello Guy_F. I thought you'd left us in disgust. I didn't *quite* call you a luddite, but I will admit I insinuated it.

Please, check my work - I want to make sure things are right. I know, we've certainly bumped heads over a few things, but I will happily take constructive criticism as long as you can help with a solution. With the caveat that it doesn't involve large amounts of concrete  ;)

Firstly, I will grossly oversimply things by considering it to be a single degree of freedom , up/down motion and then calculate a large-ish safety factor to cover my arse.

5kg is pretty much what I worked out for the reciprocating mass as well.

Taking into account a stroke of 138.7mm ... that's (reaches for calculator) :
A effective eccentricity of half that... 69mm, if you consider halfway up the bore the "centreline" and the piston oscillates about that point.
A speed of 68.1 radians/second at 650RPM
An effective force of 5kg x 0.069m x 68.1^2 = 1598 Newtons.

For the rotating imbalance forces:
Speed's still the same = 68.1 rad/s
The effective radius is the same as well.
Force is 2.5kg x 0.069m x 68.1^2 = 799 Newtons.

So a 750kg frame + both those forces added... hmmm. Are those in phase?
Mostly? I'll call it in phase for the purposes of gross oversimplification and worst-case vertical loading effects.

Anyway, 750kg + (2397N/9.81*5) = about 2000kg for a factor of 5 , similarly it's about 3100kg for a factor of 10.

Of course, those are only rough estimates, but we should be in the right ballpark with regards to the loads and the safety factor should CYA.

I was considering building a frame out of 3" thick wall box section, with similar diameter pipe as the crossmembers. Plated top and bottom - basically the torque box design that mobile_bob suggested.

Gut feeling at present tells me that such a frame, if kept relatively short, will be pretty stiff and not flex a great deal at that loading. But I'm going to try and make sure of that before I go to all the time and expense of constructing something that turns out to be inadequate. I've the means to test the frame with up to about 10 tons of static loading here at work, so I might be able to back up any guesstimated deflection numbers (or predicted catastrophic failures) with real-world figures.

From our conversations in other threads, I take it that you know a fair bit about fatigue. I don't really know that much. I'd presume that it's some sort of function of deflection and cycles, but I don't know much further. Care to expand further on it?

GuyFawkes

  • Hero Member
  • *****
  • Posts: 1184
    • View Profile
    • stuff
Re: Design of steel frame mount and vibration isolation
« Reply #39 on: October 04, 2006, 05:32:51 PM »

5kg is pretty much what I worked out for the reciprocating mass as well.

Taking into account a stroke of 138.7mm ... that's (reaches for calculator) :
A effective eccentricity of half that... 69mm, if you consider halfway up the bore the "centreline" and the piston oscillates about that point.
A speed of 68.1 radians/second at 650RPM
An effective force of 5kg x 0.069m x 68.1^2 = 1598 Newtons.

For the rotating imbalance forces:
Speed's still the same = 68.1 rad/s
The effective radius is the same as well.
Force is 2.5kg x 0.069m x 68.1^2 = 799 Newtons.

So a 750kg frame + both those forces added... hmmm. Are those in phase?
Mostly? I'll call it in phase for the purposes of gross oversimplification and worst-case vertical loading effects.

Anyway, 750kg + (2397N/9.81*5) = about 2000kg for a factor of 5 , similarly it's about 3100kg for a factor of 10.


Complete rubbish.

F = MA

Force, the thing we are after, = Mass, which we have an example number for  x Acceleration.

Since we want a peak load scenario we need to know the peak acceleration undergone by the piston.

WHat is this peak acceleration.

Show your workings.
--
Original Lister CS 6/1 Start-o-matic 2.5 Kw (radiator conversion)
3Kw 130 VDC Dynamo to be added. (compressor + hyd pump)
Original Lister D, megasquirt multifuel project, compressor and truck alternator.
Current status - project / standby, Fuel, good old pump diesel.

Andre Blanchard

  • Sr. Member
  • ****
  • Posts: 373
    • View Profile
Re: Design of steel frame mount and vibration isolation
« Reply #40 on: October 04, 2006, 06:36:47 PM »
Complete rubbish.

F = MA

Force, the thing we are after, = Mass, which we have an example number forĀ  x Acceleration.

Since we want a peak load scenario we need to know the peak acceleration undergone by the piston.

WHat is this peak acceleration.

Show your workings.

Think SIN waves, RMS, peak, 0.707 and such.


Very interesting example of fixing vibration problems in much larger engines.
Larger as in 3000 to 5000 lb weights being placed at a radial dimension of like 7 feet.

http://books.google.com/books?vid=0xWDihsZDgGVLv3SIR&id=-PwrQBRs9PYC&pg=PA529&vq=vibration&dq=Transactions+of+the+American+Society+of+Mechanical+Engineers+1905&as_brr=1

Link to PDF
http://books.google.com/books/pdf/Transactions_of_the_American_Society_of_.pdf?vid=0xWDihsZDgGVLv3SIR&id=-PwrQBRs9PYC&output=pdf&sig=ggB9DY5k2UJ0T940pGXOLqO_LNc

Title : Transactions of the American Society of Mechanical Engineers
Volume : XXVI
Copyright : 1905
Page of book : 529
______________
Andre' B

Guy_Incognito

  • Sr. Member
  • ****
  • Posts: 255
  • Just a guy, incognito.
    • View Profile
Re: Design of steel frame mount and vibration isolation
« Reply #41 on: October 04, 2006, 07:00:08 PM »
Quote
Complete rubbish.

F = MA

Force, the thing we are after, = Mass, which we have an example number for  x Acceleration.

Since we want a peak load scenario we need to know the peak acceleration undergone by the piston.

WHat is this peak acceleration.

Show your workings.

I just did, you idiot.

Radians/sec is the radial velocity. The radial velocity squared divided by the radius equals the centripetal acceleration outwards.
This is the basic formula for centripetal force.

Multiply by the mass and lo! you have F=MA and the outward force which is constantly applied throughout the entire 360 degrees of rotation on a properly rotating body. In the case of the up and down piston, it's limited to one dimension (with the gross oversimplification I mentioned before).

This is basic physics , Guy_F. It's not advanced physics, which would tell me to the nearest millinewton, but in this case - with a generous safety factor - it's quite enough.

mobile_bob

  • Administrator
  • Hero Member
  • *****
  • Posts: 2940
    • View Profile
Re: Design of steel frame mount and vibration isolation
« Reply #42 on: October 04, 2006, 07:04:03 PM »
Wow!

yup,,, religion, politics and engine mounting,,,

we just can't talk about it

bob g
otherpower.com, microcogen.info, practicalmachinist.com
(useful forums), utterpower.com for all sorts of diy info

mobile_bob

  • Administrator
  • Hero Member
  • *****
  • Posts: 2940
    • View Profile
Re: Design of steel frame mount and vibration isolation
« Reply #43 on: October 04, 2006, 07:39:14 PM »
to all interested parties

go to this link and read all that it will allow you to read

and if you want the book i guess it is available for 150 bucks or so

http://books.google.com/books?hl=en&lr=&id=jw4Xti9HGSUC&oi=fnd&pg=PR9&sig=fl0-ohbVMhPBnsRiGD-7_GIiQUM&dq=%22Rivin%22+%22Passive+Vibration+Isolation%22+&prev=http://scholar.google.com/scholar%3Fq%3Dauthor:%2522Rivin%2522%2Bintitle:%2522Passive%2BVibration%2BIsolation%2522%2B%26hl%3Den%26lr%3D

pretty fascinating to find out that good old trial and error still has a place at the table

bob g
otherpower.com, microcogen.info, practicalmachinist.com
(useful forums), utterpower.com for all sorts of diy info

Guy_Incognito

  • Sr. Member
  • ****
  • Posts: 255
  • Just a guy, incognito.
    • View Profile
Re: Design of steel frame mount and vibration isolation
« Reply #44 on: October 04, 2006, 08:33:27 PM »
Quote
pretty fascinating to find out that good old trial and error still has a place at the table

Don't get me wrong Bob, trial and error still has a place... it's just that I prefer to keep the trials to a minimum and ensure any error is as small as possible first.  For me - I will be spending a substantial amount of money on my fancy-pants isolators. Before I splash the cash, I'd really like to get as good an idea as possible of what's going to happen when I bolt it all together. If someone can use their knowledge to "trial and error" it themselves and they're using cheap mounts, that's good - but outlay costs mean I need to check things first. I happily accept that there will be fiddling, but one hopes that I will be reasonably close to the target on the first few tries.

If you're interested in a couple of good introductory texts online I can point you in the right direction - I've been all over the internet and back again over the last couple of weeks it seems. Most mount manfacturers have them on their site as well, and I suppose we could pick up the phone and actually talk to a bloke that uses and sells isolators for all sort of applications and get the good oil straight from him.... but where's the fun in that?  ;)

Actually, I have contacted a guy already, and when I tell him the sums that I've done and the mount I've picked, he tells me I'm on the right track and it should be suitable, pehaps with a little bit of damping with a shock absorber if necessary. Seeing as he works with them every day and would soon be out of a job if he gets it wrong, one can only presume that he knows what he's talking about.

And while I'm merrily blabbing to myself about what I'm doing here, I'm still interested in what everyone else is doing with their resilient mount, so don't think that I'm trying to steamroll everything.

Back to the religious debate?