Attached the schematic of the fet buffer for the accelerometer. If you can’t find a small 100M resistor for R1, a 10M one will do also. This resistor defines together with the piezo’s capacitance the low freq. cut-off, this is in fact a high-pass filter. But most piezo’s are in the range of a few nF so this is not much an issue. The circuit is very sensitive to hum, so proper shielding is necessary. Some aluminium or copper foil connected to the gnd will do.
Cheers
Cheers
Attachments
cab vibration
Mick et al - thanks.
This got me to asking myself where the vibration comes from and it sure seems like its more likely to be mechanically transmitted from the baffle than via air since the density of wood is roughly 500 times higher than the density of air.
This made me think that if you mechanically decouple the baffle from the rest of the box the cab resonances might be vastly reduced. Have people tried this? Or, is it pointless since the decoupling would have some other effect? I'm thinking something like thick rubber gasket between the baffle and the rest of the box - screwed through.
Another approach would be to limit the surface area of the baffle/wall junction (I don't know why this works - but in guitar making you definitely get better transmission by increasing the area of a joint), or to create impedements to dominant resonances by changing the density of the material at that junction.
Mick et al - thanks.
This got me to asking myself where the vibration comes from and it sure seems like its more likely to be mechanically transmitted from the baffle than via air since the density of wood is roughly 500 times higher than the density of air.
This made me think that if you mechanically decouple the baffle from the rest of the box the cab resonances might be vastly reduced. Have people tried this? Or, is it pointless since the decoupling would have some other effect? I'm thinking something like thick rubber gasket between the baffle and the rest of the box - screwed through.
Another approach would be to limit the surface area of the baffle/wall junction (I don't know why this works - but in guitar making you definitely get better transmission by increasing the area of a joint), or to create impedements to dominant resonances by changing the density of the material at that junction.
Pjotr - thanks for going to the trouble of providing all the info. Unfortunately the vibration measurement thing has got a bit bigger than I expected, and it may need to wait for a while. I have a list of other "jobs" that need to be done, with the bass speaker jumping the queue, much to the annoyance of my partner, who was expecting the magificent dining room table to be completed next. Hopefully others can be inspired to use the info a bit sooner.
Theo - Although decoupling the front baffle may seem attractive, this will probably cause additional problems. By decoupling a baffle (especially for a bass speaker), the baffle will move more than if it attached to the rest of the cabinet, and will vibrate in ways that will depend on the frequency being reproduced by the driver. In effect the baffle will be like a mass on a spring, which will have a resonant frequency, and a frequency dependent amplitude vibration.
An extension of what you are talking about is to decouple the driver from the front baffle (eg with foam or rubber), which was discussed on another thread. Same problem here - the driver becomes the mass and the foam is a damped spring.
By making it harder for the vibrations of the baffle to be transmitted to the rest of the enclosure you probably just keep the vibration duration longer in the front baffle, where it is probably more of a problem. I think that the best approach is to try to dissipate the energy as quickly as possible to a massive object (eg concrete slab floor), where it can annoy your neighbours and termites rather than you.
Mick
Theo - Although decoupling the front baffle may seem attractive, this will probably cause additional problems. By decoupling a baffle (especially for a bass speaker), the baffle will move more than if it attached to the rest of the cabinet, and will vibrate in ways that will depend on the frequency being reproduced by the driver. In effect the baffle will be like a mass on a spring, which will have a resonant frequency, and a frequency dependent amplitude vibration.
An extension of what you are talking about is to decouple the driver from the front baffle (eg with foam or rubber), which was discussed on another thread. Same problem here - the driver becomes the mass and the foam is a damped spring.
By making it harder for the vibrations of the baffle to be transmitted to the rest of the enclosure you probably just keep the vibration duration longer in the front baffle, where it is probably more of a problem. I think that the best approach is to try to dissipate the energy as quickly as possible to a massive object (eg concrete slab floor), where it can annoy your neighbours and termites rather than you.
Mick
Jonathan:
I do in no way doubt that it would work the way you described it. I just read a test (verdict: "State of the Art 😎 ") of your mono cartridge the other day, and thought it might be expensive to take one of your cartridges for this purpose (with the mono one it wouldn't even work at all) !
I do fully agree that it doesn't have to be perfectly flat in terms of amplitude response. I assume that the FR of panel resonances must show very extreme peaks and dips anyway.
Regarding the stroboscope: I remember that I once read in "Speaker Builder" that someone did it like that. But I don't have any details. I did some thinking and came to the following conclusion how I would do it:
Instead of a xenon-flash I would take a superbright LED (or even multiples) controlled by a one-shot. The trigger circuit must be able to trigger on the signal peaks AND the zero crossings (user selectable) because the resonances do not necessarily have to be in phase with the excitation. What the strobe then does is making the flexing visible.
I.e. it would show you the pattern of the flexing but not the amplitude like an accelerometer does.
Regards
Charles
I do in no way doubt that it would work the way you described it. I just read a test (verdict: "State of the Art 😎 ") of your mono cartridge the other day, and thought it might be expensive to take one of your cartridges for this purpose (with the mono one it wouldn't even work at all) !
I do fully agree that it doesn't have to be perfectly flat in terms of amplitude response. I assume that the FR of panel resonances must show very extreme peaks and dips anyway.
Regarding the stroboscope: I remember that I once read in "Speaker Builder" that someone did it like that. But I don't have any details. I did some thinking and came to the following conclusion how I would do it:
Instead of a xenon-flash I would take a superbright LED (or even multiples) controlled by a one-shot. The trigger circuit must be able to trigger on the signal peaks AND the zero crossings (user selectable) because the resonances do not necessarily have to be in phase with the excitation. What the strobe then does is making the flexing visible.
I.e. it would show you the pattern of the flexing but not the amplitude like an accelerometer does.
Regards
Charles
- Status
- Not open for further replies.