NVH and natural frequency

[neongreen]

I need some help understanding an NVH issue. I swapped an older inline 4 (Ford 2300 turbo, no balance shafts) into a 1962 Ford Falcon. This engine was originally in a Ford Ranger, and while it was never a "smooth" engine, it wasn't objectionable either.

When I put it into the Falcon, I immediately noticed a huge jump in vibration coming through the body of the vehicle. A number of things were changed in during the swap process. Intake manifold, exhaust manifold, fuel system, turbocharger, camshaft, ignition system, engine management, were all changed in addition to a larger clutch and accompanying aluminum flywheel, as well as mounting system. I thought that the most likely culprit would be the vibration isolators in the motor mounts, since they were very small, hard urethane bushings. I remade the mounts to accommodate a larger OEM style hydraulic fluid filled isolator.

comparison of vibration isolators

After that change, the vibration transmitted into the body structure was reduced dramatically, but there were still two problem areas, both of which appeared to be RPM dependent. It's probably easier to show rather than tell what I'm observing, so I'm attaching a sketch of what I've observed. I don't have actual data at this point, but hope to gather some in the future. When at its highest amplitude, the steering column visibly shakes, and change in the ash tray will rattle loudly. It is quite unnerving in fact.

sketch of vibration amplitude vs RPM

As you can see, the big issue is the vibration amplitude peak at around 3800 RPM. Obviously the vibration is coming from the engine, but why is there a peak in the amplitude? Are the vibration isolators insufficient? Is it due to the lack of structural rigidity in Ford's first attempt at unibody construction? Should I be looking at the possibility of an off-balance flywheel or clutch? What do I need to understand about vibration and natural frequencies to solve this problem?

 

[VEBill]

"...a larger clutch and accompanying aluminum flywheel..."

Others would know for sure, all I can us ask.

Was the original flywheel helping to balance the engine? 'Externally Balanced'

This sort of thing HERE.

 

[VEBill]

Here's another interesting link.

https://www.hemmings.com/magazine/hmn/2014/07/1965-and-Newer-Ford-Flywheels/3740311.html

 

[neongreen]

The original is internally balanced (no weights on the flywheel) and I can't verify, but I don't recall seeing any weights on the aluminum flywheel. I'm pretty certain it as a zero-balance unit.

So for the sake of argument, lets assume that an improper flywheel is not the issue.

----------------------------------------
I could get so much done if I didn't have to go to work

 

[pontiacjack]

Random thoughts: Flat-crank inline engines have a 2nd order imbalance that the Falcon six did not have. Taming it has caused much development (and many balance shafts implemented) by auto manufacturers. They weren't very successful with the '63 Tempest four I had. Perhaps some portion of the Falcon structure has a resonance near 127 Hz ([3,800 RPM x 2]/60). I could believe it- I've owned a dozen early Falcons, some of which had terrible NVH issues from other disturbances- suspension, tires, whatever.
Does your '62 have the leaf spring mount at the rear of the transmission? If so, I'd try changing it (eliminate the springiness).
Good luck. Keep us posted.

 

[Tmoose]

Hi Pontiacjack ! Haven't heard from you for quite a while. Hope all is well.

regards,

Dan T

 

[neongreen]

Hey Pontiacjack,

I think you hit on what I'm suspecting right now. I'm thinking that the interface between the front subframe and the rockers is really floppy. All of the forces and vibrations on the front subframes (including motor mounts and all the vibrations that go along with that) must go through the firewall and floor, neither of which are terribly rigid.

To increase the rigidity of the body I did install subframe connectors from the rear of the front subframes to the front of the rear subframes, but that doesn't address the worst issue in the unibody. In later first-gen Falcons and Mustangs Ford went to the trouble of adding torque boxes as I'm sure you know. These were to tie the rockers and subframes together with another method besides the not-so-rigid 18 gauge sheet metal in the floor and firewall.

After what you said, and a conversation with a coworker who has had to deal with NVH issues in a production vehicle, here is what I think is happening: The natural frequency of the firewall area coincides with the second order frequency of the 4-cylinder engine near 3800 RPM. The motor mounts can't absorb the vibration sufficiently, and when those frequencies line up, that area of the body structure resonates at an amplitude that is really uncomfortable. From about 4000-7000 RPM, there are no parts of the body structure with a natural frequency which corresponds to the frequencies generated by the engine.

Therefore, I must modify the body structure in such a way that it increases the natural frequency of the subframe/firewall interface beyond the frequencies generated by the engine. By installing torque boxes, I should be able to in a way triangulate that area of the body structure, and tie it to the rockers.
this will stiffen this area, and hopefully move the natural frequency of the body out of the range of frequencies generated by the engine.

Does that sound right?



----------------------------------------
I could get so much done if I didn't have to go to work

 

[pontiacjack]

I really don't know- I was just tossing out some thoughts.
Did you really connect the subframe connectors to the "... rear of the front subframes..." as you said? Here's why I ask: my daily driver '67 Falcon is badly rusted, especially the torque-box areas, although all four subframes are still sound. So I bought subframe connectors (some guy in Texas) but had them made longer than normal, to weld about 16" of them to the sides of the front subframes. This, in combination with a pair of stiff braces from the firewall down to the subframes, was able to correct the terrible sag in the car (jacked up the rear of the subframes enough to "straighten" the car prior to welding). [For some reason, Ford attached the '67 subframes to NOTHING behind the firewall- they merely provided a rear transmission mount.] Perhaps your subframe connectors should overlap more of the front subframes? [By the way, the improvement in my car's steering/handling/tightness/quietness/suspension-composure/etc. was amazing.]
What does your Falcon's transmission mount consist of?

 

[neongreen]

Transmission Mount

Connection to Front Subframe

Underbody Prior to SFCs and Transmission Mount

Location of SFCs

As you can see, the SFCs are well behind the engine compartment, so I don't think they are contributing a lot to the stiffness of the firewall area. They are stiffening up the middle of the car, and if anything are exposing that firewall interface as the weakest link. This is why I think torque boxes, or something similar could improve the area between the SFCs and the suspension crossmember.

----------------------------------------
I could get so much done if I didn't have to go to work

 

[pontiacjack]

Thanks for the pictures. Let us know the outcome.

 

[Panther140]

I think you're on the right track with the root cause and solution to this problem.

My advice, take a strobe light and set it to whatever the natural frequency of your firewall is. Point it at the firewall. Then, sweep the engine through the RPM range. What you see at 3800 RPMs will be your firewall appearing to flex/warp in slow motion.

"Formal education is a weapon, whose effect depends on who holds it in his hands and at whom it is aimed." ~ Joseph Stalin

 

[bugbart]

Hi Neon Green
How did you go with your NVH issue? Have you resolved it yet?
If not do you mind me asking what type of gearbox is attached to the engine, and do you have a DMF or SMF?
I might be a bit behind in this forum forgive me if my questions have been asked before.
The ranger itself has many fixes for NVH issues, e.g. cab isolators and transmission damper for powertrain bending modes.
A steering column damper may help as well.
Please keep us posted on the problems and solution as I intend to put an I5 ranger engine in falcon BA .

 

[neongreen]

Panther, that's an ingenious way to check. I suppose I could make my timing light do that for me.

Bugbart, I have not yet had time to attempt to fix it. I'm not sure what you mean by SMF vs DMF... but it has a 5 speed manual transmission. I'm sure a column damper would help, but I'm not thinking it's the main offender at this point.

----------------------------------------
I could get so much done if I didn't have to go to work

 

[BrianPetersen]

"SMF" = "single mass flywheel", "DMF" = "dual mass flywheel". Has to do with where the vibration damping part of the clutch is. With a SMF the flywheel is one big chunk of steel and the friction plate has springs in it oriented in the direction of torque transmission, that drive the input shaft of the transmission. With a DMF that contraption is built into the flywheel itself (thus, the flywheel has "dual masses" - one part bolted to the crankshaft and the other part isolated from torsional vibrations) and the friction plate is splined directly to the transmission input shaft - no springs in it.

I think the Ford Pinto engine was developed before the DMF was invented. Unless someone has done some exceptionally odd parts swapping, you will have a SMF and a normal clutch.

 

[neongreen]

Ahhh OK. I had not heard that terminology before. Yes, it's definitely a single mass flywheel.

----------------------------------------
I could get so much done if I didn't have to go to work

 

[GregLocock]

Pinto would not have had DMF for 3 reasons - cost (I'm tempted to say cost, cost), probably not invented then, and RWD I4 don't really need them NVH wise until ca 1995. They do help a little bit with some things but really not appropriate for a nasty little econobox.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[neongreen]

*********** FOLLOW UP POST **********

I was not able to find any pre made torque boxes, so I made my own.
The process can be seen in the links below.

To summarize, I'll post an excerpt from the second blog post:

"Did the torque boxes change the natural (or resonant) frequency of the subframe / rocker interface, thereby reducing a disconcerting vibration in the body structure when the engine is around 3800 RPM? The short answer is Yes.

I test drove the Falcon briefly and took it through the RPM range a few times. When I hit the problem RPM range, there was still more vibration than at other RPM ranges, but the amplitude was greatly reduced. While is still more than my ideal situation, but I have to remember that this is a shaky old engine in a shaky old car, and that it will likely improve once sound deadening materials and an interior are installed. My initial feeling is that my DIY Falcon torque boxes are doing the job I hoped they would do."

http://ironhydroxide.blogspot.com/2017/04/boxes-of-torque.html

http://ironhydroxide.blogspot.com/2017/05/boxes-of-torque-part-dos.html

----------------------------------------
I could get so much done if I didn't have to go to work

 

[pontiacjack]

Nice work. Thanks for posting the patterns. Glad your persistence paid off.

 

[Strong]

Your structural work is impressive. Did you try the original steel flywheel which should change the torsional natural frequencies and be different from the aluminum flywheel?

Walt