• TalkBass has been independent since 1998. Add your voice.
    Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
    Join free Log in
    Want zero display ads or expanded classifieds tools? Compare plans.

Double Bass Good vibrations???

#42
Matthew Tucker said:
I'll be cutting my FFs soon too. laid them out last night and decided i want them slightly larger; the bass I have made has turned out 2-3cm longer than the original I was copying, not sure why, and it doesn't matter really.

I tried supporting the plate at the thirds points as you suggested, but that doesn't work very well,as the lower bout is heavy and tends to tip the whole plate up. might have to weight the neck cutout area! And the tap tones seem pretty much the same all over, maybe a semitone difference in places but its hard to tell. I get a more solid thunk just next to the blocks, but that's because i have left those areas thicker, my hunch is that by leaving the thicker centre strip thicker the top will be less likely to collapse longways over time, but allow the top to vibrate laterally.
Click to expand...


Matt, can you try the chladni test after the ff and bassbar if you get a chance. See if you can then see anything that looks like a mode 2 around 40 Hz and mode 5 around 80Hz. and get the weight also.
Good on ya mate!
 
#43
Well this one is interesting; It doesn't correspond to any of the classical violin modes.

http://www.phys.unsw.edu.au/jw/patterns1.html

114Hz
Invalid Link Removed

this may be on the way to a mode 2 at 36Hz
Invalid Link Removed

and this is what I get at 70Hz. nothing again until 91Hz

Invalid Link Removed

91Hz
Invalid Link Removed

It doesn't really seem to matter where I place the foam blocks. I tried many positions, but I get the same patterns more or less.
 
#44
Looks good!

36 hz is mode 2 related to your cross grain stiffness and 70 hz is mode 5, your long grain stiffness. Get the plate weight at this point.

Also the mode 5 pattern is telling you that the the upper bout area at the very top of your plate is stiff and won't bend. It should show a straight line across the plate there if the plate is flexing in that area. Don Harris emailed me some of his photos where he used that mode to thin that very area and bring the line down.
 
#45
Dunno ... 70 Hz looks very like 36Hz to me. The only nodal lines created are the lateral ones , the one going across at the bottom was more or less swept there by me, the pattern across the upper bout is real but I would'nt know where to start thinning. the upper bout IS a bit less flexible longways than the lower bout, but the tap tones are just as clear as the lower bout. I'll have a few thinks.
 
#46
Got some clearer patterns I think this morning. Obviously I didn't take the Marilyn Monroe mode far enough ...

I decided that suspending the plate at the edges prevented too much vibration, so I am resting the bridge area of the plate on a piece of foam and allowing the whole thing to pump. The centre of the plate doesn't flex much anyway due to the arching.

The only way I can get the upper bout to move is to excite the lower bout hard.

Ken, it seems that where mode 5 is working on the lower bout, mode 2 is active in the upper!

Invalid Link Removed

Invalid Link Removed

The ring you can see is where the rubber of the speaker touches the plate when I get too close!
 
#50
Glad to see you are still interested Arnold, even if it's just for some cheap thrills.

Matthew, Post number 43 shows the modes much clearer to my eye. But of course I am not there watching the sand dance. Supporting the plate where it is vibrating the most, like where the bridge is, might be telling you something but I don't think it is going to help find the 2 and 5.

Quite frankly I didn't think you would be this persistent...well, you need to support the plates where the sand sits in a node line. For the mode 2 (36 Hz) it is where the sand almost runs up to the flat edge on each side of the upper bout and same on the lower bout. If sand is dancing, brush it off the plate so it won't show on the photo when you turn it off. For mode 5 try to put blocks under the node line where the corners would be if it had them but in where the sand is. Metal glitter or tea leaves or poppy seads work better because they are lighter.

If you are not able to do any further testing, if the neighbors are calling the police, the dogs are leaving town, the kids are about to kill you or Arnold has a heart attack, I would be satisfied that your mode 2 is 36 Hz and 5 is 70. Did you get the weight yet?
 
#53
Here's my best Mode 2, supported at the nodal lines, at 33Mhz

Invalid Link Removed

But Ken, I can't get a mode 5 going with foam blocks under the conventional "corners" maybe cos I don't have any?

with the plate balanced on two single foam blocks across the middle - where it balances - I get this at 91Hz. Could this be an elusive mode 5? I think the plate may be very stiff longways still, and I can't get the mode 5 with the speaker setup I have. Might need a more powerful amp.

Invalid Link Removed

... and the plate weight is exactly about 1500 grams!

What have I learnt so far?

DON'T use coffee grounds :(
 
#55
Who hooo! This is like fishing! :hyper: Brush it all into a pot and make some strong coffee.

That is a clear mode 2, that is good. I though that would be the hard one to get.

Mode 5 is not showing up as I would have thought. Well 91 Hz might be it. If it is, the plate is variably stiff in the long grain direction.

Keep in mind that I do not propose this as a help (yet) for you to proceed with graduating the plate, just measuring what is.

If you have the fortitude and coffee, once you get the ff's cut and the bar on, then do it again and see what happens.
 
#56
Well ... it might be that a cornerless shape doesn't have the same mode 5 as a violin shape. Or it might be that Cedar shows modes differently to Spruce.

Another series of pictures shows one corner of the upper bout is significantly different to the other one. That gave me good pointer, I think, to trying to work out why. And I did find a very thick area I hadn't picked up.

What have you managed to get? got any pics?

by the way, I ran out of tea one night, used coffee ... and it was a bad move. coffee ground too fine and stuck to everyhing and made the workshop smell like three day old coffee grinds. Tea leaves are much better.
 
#57
Very interesting, thanks for sharing. I have a question.
What is the relevance of the Chaldni patterns? I mean, I know they represent the eigenvalues for the vibrational spectrum of the top; however, you will be changing the boundary conditions from a (quasi)-free plate to a (quasi)-bound plate (or at least pinned edges) and the eigenvalues will be displaced to higher frequencies.
Cheers
David
 
#58
Well, as we were discussing earlier in the thread ... we (Ken M and I) are interested in trying to use the patterns to measure or depict some properties of the free plate, that's all really. Like measuring the amount of flour in a cookie recipe; whether you measure in grams or ounces or cups or packets or gumboots or by eye ... is not relevant. But unless you can find a way of measuring the amount of flour, you can't really repeat or communicate the recipe.

The frequency the patterns occur at are, to my mind, not relevant at all to the sound or resonance of the final instrument. they are just numbers. Is it not reasonable to assume that two plates that exhibit similar patterns at similar frequencies might behave in a similar way when attached to the instrument?

Whether that is a fruitful exercise or not is open to debate. We're experimenting :)

(edit: it is also possible to get chladni patterns off a finished instrument. I think this is where some people try to use the patterns to tune the sonic performance of the instrument. But i have no idea about this and only a passing interest ... at this stage!)
 
#59
DavidMP said:
Very interesting, thanks for sharing. I have a question.
What is the relevance of the Chaldni patterns? I mean, I know they represent the eigenvalues for the vibrational spectrum of the top; however, you will be changing the boundary conditions from a (quasi)-free plate to a (quasi)-bound plate (or at least pinned edges) and the eigenvalues will be displaced to higher frequencies.
Cheers
David
Click to expand...
Dr. William Atwood, a physicist and violin maker from Santa Cruz CA, did research on fixed plate tuning and how it relates to free plate tuning. He did a presentation for the Catgut Acoustical Society and wrote a research paper for the CAS Journal on his findings. Unfortunately, the paper is not available to read other than in the printed journal. Sorry, but I can't give any specifics since all of my CAS Journals are currently packed away in storage. Perhaps another CAS member here can look up when the article was published if anyone is interested.
 
#60
I tried rig something up to glitter test my bass top plate but was really unable to get it going. It is hard to even get a good sound at 80 hz let alone 40! What ampage are you using? I see you have a PC to generate the tone. Are you using a sub woofer? I have seen cheap, powered subwoofer on ebay and wondering if I should pick one up.

Mostly I have used tap tones to get what I think are the mode 2 and 5 tones. But of course it is not as clear as the glitter patterns.

My method: Hold the plate at the neck end of the plate where mode 2 and mode 5 would have a common nodal line. I do this by feel, I try to feel for vibration with the fingertips. It is pretty hard on a plate the size of a bass, but possible. I thump the plate with a piano hammer (felt tip) at different areas to excite the plate while pointing it at the mic. Using a sound analysis program I record the thumps and then analyze them with FFT analysis. This gives peaks at frequencies in the tap.

For my top plate with ff's cut and graduation only about 80% finished. Results are: peaks at 40 Hz (very low, maybe not real), 79 Hz, 150 Hz and 210 Hz.

My wish list would include a mic that can measure 30 hz I wonder if using a speaker as a mic would actually work better at these low frequencies? Any engineers out there?

The reason I test my free plates is simple. My bass will sound good if it sounds good in the end and bad if it sounds bad! No amount of testing can change that. The proof will be in the pudding. I am testing because I might want to use this test info to try to do something again that is similar. I have data from previous tops that I tested and 1 with wood from the same tree. I think it is also a good idea to compare with others.
 

Latest posts

Top Bottom
Back