I am designing a Parallel Push Pull 300B Amp using some fairly expensive components such as Plitron Power and Output Trannies.
I had an idea of mounting the main tube circuitry on a sub-chassis isolated from the main chassis via resilient mounts, so started to look at the basic theory.
I was quickly apparent that if you don’t know what you are doing (and at this stage I don’t) it is quite easy to make the problem of coupled mechanical vibration worse rather than better.
Given:
fe = excitation frequency
fn = natural frequency (resonant frequency) of the resilient mount
If fe/fn < root 2 then you get amplification of the vibrations rather than attenuation
For good attenuation you need fe/fn > root 2 (much greater is better).
Now here is where I need help:
1) Can I assume that the mechanical excitations are mainly from the power transformer and excitation frequency will basically be 50Hz and 100Hz
2) When I start looking at the various resilient mounts (Ettinger Rubber Resilient Mounts is what I looked at) they do not generally give a figure for the natural frequency but simply give a value of the Shore Hardness of the rubber material (options of 43, 57 or 68).
3) Can I conclude anything useful (such as resonant frequency) from the Shore Hardness values. Is there a table for various materials cross referencing these things?
Any guidance gratefully received.
Cheers,
Ian
I had an idea of mounting the main tube circuitry on a sub-chassis isolated from the main chassis via resilient mounts, so started to look at the basic theory.
I was quickly apparent that if you don’t know what you are doing (and at this stage I don’t) it is quite easy to make the problem of coupled mechanical vibration worse rather than better.
Given:
fe = excitation frequency
fn = natural frequency (resonant frequency) of the resilient mount
If fe/fn < root 2 then you get amplification of the vibrations rather than attenuation
For good attenuation you need fe/fn > root 2 (much greater is better).
Now here is where I need help:
1) Can I assume that the mechanical excitations are mainly from the power transformer and excitation frequency will basically be 50Hz and 100Hz
2) When I start looking at the various resilient mounts (Ettinger Rubber Resilient Mounts is what I looked at) they do not generally give a figure for the natural frequency but simply give a value of the Shore Hardness of the rubber material (options of 43, 57 or 68).
3) Can I conclude anything useful (such as resonant frequency) from the Shore Hardness values. Is there a table for various materials cross referencing these things?
Any guidance gratefully received.
Cheers,
Ian
1. I would be more worried about air borne vibration at higher frequencies and structural vibration at very low frequencies. Suppression will depend upon the amp + support structure and not just the tube mount.
2. They cannot have a "natural frequency" given as it is a function of the carried mass as well. Not sure if calculating this will be of much help anyway. Successful vibration control is the result of a combination of treatments, usually isolation and damping and certainly a lot of listening.
The durometer will give you an idea of the spring constant (translating that to n/cm or anything useful is not straightforward!) but will tell you nothing about the loss (in this context, a high loss is a good thing). I'd go as soft as I could find, then use mass and damping to empirically optimize.
For a low level circuit (e.g., a phono stage), this is FAR more critical. If you have significant microphonics in a power amp, you've done something terribly wrong.
For a low level circuit (e.g., a phono stage), this is FAR more critical. If you have significant microphonics in a power amp, you've done something terribly wrong.
Hey Ian,
I suggest you experiment and test some home-cooked suspensions. The commercially available ones I´ve found seems to be far to hard. I am just about to try a Casco/Nobel silicone rubber for glass. It is the softest one I´ve found and is said to be 16 Shore A. My aim is to use them for a really low level DHT-preamp PCB.
I suggest you experiment and test some home-cooked suspensions. The commercially available ones I´ve found seems to be far to hard. I am just about to try a Casco/Nobel silicone rubber for glass. It is the softest one I´ve found and is said to be 16 Shore A. My aim is to use them for a really low level DHT-preamp PCB.
I have seen shock mounts used often in small helicopters to isolate avionics black boxes from the airframe, they are rated in weight ranges for the mass they will support, anything from 1/2 kg to a few kg. I have not seen them at all in larger aircraft. I don't know what's inside but they feel "dead" and have around 1" travel. Be interesting to dissect one. I could have given you some when I worked on choppers, but that was quite a few years ago. It's possible light aircraft use them too. Might be something to look at. Since aircraft are continually being modified to keep up with Federal aviation standards, there should be quite a few racks with shock mounts sitting around gathering dust that will never be used again.
Not the first search-phrase that came to me😉. Thanks a lot, great site! Hopefully they sell small quantities too.
What was that stuff used in speakercone surrounds?....Was it called 'Nomex'??? Can't recall too clearly.......I must reread/scan my textbooks thru & thru. I think that was the name. Surrounds must damp out any & all frequencies to be encountered, lest they reradiate back into the cone.
_______________________________________________Rick.........
_______________________________________________Rick.........
Lord Isolators
Lord Corporation is a big player in the vibration isolation business. They have mounts from tiny to gigantic in lots of form factors.
Vibration, Shock & Motion Control Products
In my experience, the low profile "motor mount" style isolators give better isolation than the grommet style isolators.
http://www.lord.com/Home/ProductsSe...ProfileMountsAMSeries/tabid/3476/Default.aspx
Lord Corporation is a big player in the vibration isolation business. They have mounts from tiny to gigantic in lots of form factors.
Vibration, Shock & Motion Control Products
In my experience, the low profile "motor mount" style isolators give better isolation than the grommet style isolators.
http://www.lord.com/Home/ProductsSe...ProfileMountsAMSeries/tabid/3476/Default.aspx
Last edited:
"Norsorex" a trademarked material used in surrounds.
Perhaps a small quantity of this stuff will damp out those pesky vibrations.
___________________________________________________Rick....
Perhaps a small quantity of this stuff will damp out those pesky vibrations.
___________________________________________________Rick....
Mass and stiffness is your friend here as we are dealing with mostly "low frequencies".
I had an old CJ 'Nuvistor based' head amp that used rubber bands. It worked pretty well until the rubber bands dried up and broke. We also used to use sorbothane, but it would eventually flow and not be effective.
Determining the nodal frequency is best done by experimentation unless you have access to some good modeling software. Easy enough to do with a sound card spectrum analyzer.
I had an old CJ 'Nuvistor based' head amp that used rubber bands. It worked pretty well until the rubber bands dried up and broke. We also used to use sorbothane, but it would eventually flow and not be effective.
Determining the nodal frequency is best done by experimentation unless you have access to some good modeling software. Easy enough to do with a sound card spectrum analyzer.
Ian: As mentioned the fn is a function of your suspended mass and the stiffness of the mounts (usually x 4 mounts).
One empirical method involves mounting an accelerometer on the suspended mass and whacking it with an accelerometer equipped hammer and looking at the response/attenuation, or better yet, exciting it with a 50Hz/60hz signal of varying amplitude through a vibration shaker.
You could fab a small shaker rather easily with a small dc motor with a weight mounted eccentrically on the shaft spinning at 3000 rpm to sim 50 hz excitation. A trip to the adult toy store and some reverse engineering might be in order here........ Mount the shaker on the chassis and compare the amplitude response of the chassis to the amplitude response of the suspended mass with an accelerometer.
Do iphones have accelerometers in them? They seem to have everything else........
Several practical considerations when sizing isolators:
You typically want the most compliant mount that will support the load (fe/fn > root 2 ), but super-soft mounts won't support much weight without excessive sagging. They are also prone to failure when bumped hard or dropped. You can use travel-limiting washers to prevent this failure mode with the mount style shown below. (The travel limiting washers are bottom mounted.)
Grommet style mounts are easy to use, but are in general stiffer than the motor-mount style or the elastomeric block (sandwich) style.
Grommet style: McMaster-Carr
bolt down style: McMaster-Carr
sandwich mount style: McMaster-Carr
I have bunch of Lord isolator mounts as shown below. If you want a handful to experiment with, PM me your address and I'll send some over.
One empirical method involves mounting an accelerometer on the suspended mass and whacking it with an accelerometer equipped hammer and looking at the response/attenuation, or better yet, exciting it with a 50Hz/60hz signal of varying amplitude through a vibration shaker.
You could fab a small shaker rather easily with a small dc motor with a weight mounted eccentrically on the shaft spinning at 3000 rpm to sim 50 hz excitation. A trip to the adult toy store and some reverse engineering might be in order here........ Mount the shaker on the chassis and compare the amplitude response of the chassis to the amplitude response of the suspended mass with an accelerometer.
Do iphones have accelerometers in them? They seem to have everything else........
Several practical considerations when sizing isolators:
You typically want the most compliant mount that will support the load (fe/fn > root 2 ), but super-soft mounts won't support much weight without excessive sagging. They are also prone to failure when bumped hard or dropped. You can use travel-limiting washers to prevent this failure mode with the mount style shown below. (The travel limiting washers are bottom mounted.)
Grommet style mounts are easy to use, but are in general stiffer than the motor-mount style or the elastomeric block (sandwich) style.
Grommet style: McMaster-Carr
bolt down style: McMaster-Carr
sandwich mount style: McMaster-Carr
I have bunch of Lord isolator mounts as shown below. If you want a handful to experiment with, PM me your address and I'll send some over.
Attachments
Last edited:
McMaster-Carr
I haven't found flow to be too much of a problem. I use it in a phono amp, and to isolate my tt motor from the turntable. Works like a charm.
Sheldon
I haven't found flow to be too much of a problem. I use it in a phono amp, and to isolate my tt motor from the turntable. Works like a charm.
Sheldon
Well.............depending on the actual mass, do not forget the good old fashioned 3hz spring approach used in the AR turntable.
I once had some really hyper pooged ARC D-76A's where one pulled the power supplies into separate cabinets, and suspended the audio signal circuits (properly weighted of course) with those springs which were then avail. by the truck load..
'Course we also had a whole avionics lab with hundreds of different resonant freq. shock mounts to play with. YES...you can hear the difference.... and yes, rubber band (or "o" ring) suspensions work very effectively. BUT you must first know the mass to be damped and then the natural resonant freq. of that mass AND the masses and natural res. freq. of the base one is mounting ( that is transmitting vibration) into or decoupling from. 3 point suspensions often are best. Otherwise slightly off tuned 4 & 5 point suspensions absorb/decouple the load very very well. Golly... Miss Molly..... then there are viscous fluid dampers....and Butyl tape, Silly Putty....... the sky is the limit!
Have fun and enjoy the journey.
By the way Lord very likely does have a set of nomographs and charts to equate shore/durometer values to natural resonant freq. We often used such to calc. engine to mount vibration dampening -coupling when designing aircraft engine mount weldment fatigue life.. etc.
Time for me to shut-en-ze-mouth.
I once had some really hyper pooged ARC D-76A's where one pulled the power supplies into separate cabinets, and suspended the audio signal circuits (properly weighted of course) with those springs which were then avail. by the truck load..
'Course we also had a whole avionics lab with hundreds of different resonant freq. shock mounts to play with. YES...you can hear the difference.... and yes, rubber band (or "o" ring) suspensions work very effectively. BUT you must first know the mass to be damped and then the natural resonant freq. of that mass AND the masses and natural res. freq. of the base one is mounting ( that is transmitting vibration) into or decoupling from. 3 point suspensions often are best. Otherwise slightly off tuned 4 & 5 point suspensions absorb/decouple the load very very well. Golly... Miss Molly..... then there are viscous fluid dampers....and Butyl tape, Silly Putty....... the sky is the limit!
Have fun and enjoy the journey.
By the way Lord very likely does have a set of nomographs and charts to equate shore/durometer values to natural resonant freq. We often used such to calc. engine to mount vibration dampening -coupling when designing aircraft engine mount weldment fatigue life.. etc.
Time for me to shut-en-ze-mouth.
Last edited:
exactly! You may try a local Avionics shop as they often have mounts pulled from otherwise surplussed equipment.
I got good results using sheet neoprene - the wet suit stuff - to isolate a noisy TT motor if you want to DIY. Check with upholstery/surf shops. Morgan Jones suggests narrow elastic like in women's undies to make sort of a trampoline.
I've successfully used mini-bungee cord to suspend hard drives in my PC, works excellent from a noise standpoint until the drives die an early death from overheating.
- Status
- Not open for further replies.