OK, so I have 200 2in cheap full range drivers on the way (yes the Parts Express buyout ones). I was thinking of experimenting with line arrays with 3 columns of drivers 33-34-33 each side with the center column offset half a driver length.
I'll till try this configuration anyway out of curiosity.
I'm sure people have tried planar arrays with conventional drivers?
How do you think a 10x10 array would go? maybe convex curved? dipol?
I'm not expecting great sound from $2 drivers but with 200??
Or am I just being silly?
Cheers Ben
I'll till try this configuration anyway out of curiosity.
I'm sure people have tried planar arrays with conventional drivers?
How do you think a 10x10 array would go? maybe convex curved? dipol?
I'm not expecting great sound from $2 drivers but with 200??
Or am I just being silly?
Cheers Ben
You'll be able to tell us after you fire them up
200 holes!!!, 400 terminals to solder!!!
This one?
An externally hosted image should be here but it was not working when we last tested it.
dave
I think it would be great! A friend of mine experimented with 4x4, the results were very promising. But be prepared to use active equalization. It is really needed to get any bass and also highs (because outputs of individual drivers combine up to a certain frequency only)
line arrays with multiple small drivers are out there
on the other hand 10x10 would really something new, a significant contribution to DIY audio
it has potential to become Quad ESL on steroids
and there is more - with 100 per side you can wire them all in series to get multikohm speaker for a different of kind amplifier like SET without OPTs for example
how about that!
best regards,
graaf
Last edited:
thousand drivers or thousand ohms?
the driver is 8 ohm nominal so 100 makes 800 ohm, pretty close to thousand
I am no expert on tubes but wouldn't some triodes be quite happy with 800 ohm load? Would be not enough for 300B for sure but what about 6C33C?
best,
graaf
Two columns of hundred drivers
Build two tight columns, 50 drivers each. Attach them into holes that you could drill to a plastic tube. If your room height does not allow that, leave out few drivers. You could build two such sets of speakers.
Build two tight columns, 50 drivers each. Attach them into holes that you could drill to a plastic tube. If your room height does not allow that, leave out few drivers. You could build two such sets of speakers.
Last edited:
Parafeed without a transformer
That would be a parafeed amplifier without the usual output audio transformer. Sounds interesting.
the only way to go is in a long single line source... take a look at my website under Archives, then CVSR. that's a line source from 1976 or so made up of Peerless 5.25" drivers.
Although it is not immediately obvious, the wider the baffle you use, the better the sound will be.
Pay attention to diffraction issues.
You will need to wire them in series parallel to get to a reasonable impedance.
zero gap between drivers.
There are several nice benefits from multiple drivers:
- lots of power handling
- able to EQ the bottom end
- distortion is reduced as the power per driver for a given SPL is dropped, divided by the number of drivers.
- depending on how many are in parallel the sensitivity is increased compared to a single driver
things like the multiple driver schemes posted above make for a very irregular pattern due to multiple interference effects, probably not a great idea.
There are vertical lobing issues as the freq goes up with a line source, but you can likely live with them or else roll off the array before they get objectionable and add in something on the HF above that point.
_-_-bear
Although it is not immediately obvious, the wider the baffle you use, the better the sound will be.
Pay attention to diffraction issues.
You will need to wire them in series parallel to get to a reasonable impedance.
zero gap between drivers.
There are several nice benefits from multiple drivers:
- lots of power handling
- able to EQ the bottom end
- distortion is reduced as the power per driver for a given SPL is dropped, divided by the number of drivers.
- depending on how many are in parallel the sensitivity is increased compared to a single driver
things like the multiple driver schemes posted above make for a very irregular pattern due to multiple interference effects, probably not a great idea.
There are vertical lobing issues as the freq goes up with a line source, but you can likely live with them or else roll off the array before they get objectionable and add in something on the HF above that point.
_-_-bear
the only way is a religious thing
but yes - line source like IDS25 is ok as such
to have as much drivers wired as possible (with this high combined impedance in mind) perhaps it could be made back to back push-push bipole or push-pull dipole arrangement
indeed such arrangements are highly directional, sweet spot is really just a spot, but in this sweet spot there are no problematic irregularities due to multiple interference
as I posted before a friend of mine had 4x4 tested and measured
at normal listening distance of ca 3m sound wave path length differences are not big enough to create audible problems
You can see measurable problems at standard measurement distance of 1m but there are no audible problems at standard listening distance of ca 3m
it cannot be excluded that these are really issues for microphone only but not for our sense of hearing
measurable problems but not audible problems
best regards,
graaf
my suggestions:
in case of 10x10 and similar - horn supertweeter on top of a "QuadESL-like"-looking shallow box
in case of a 1x50 line source - only a ribbon or magnetostat with restricted vertical dispersion to match the rest of the line
best regards,
graaf
This is a nice idea - but not quite correct.
the large tall vertical array - especially if it happens to be the height of the room does not exhibit any restriction on vertical dispersion anywhere in front of the speakers... to the contrary, it is exceptionally even at frequencies below the 1/4 wave of the inter -driver center-to-center distance. Even at frequencies above that point the variations in HF response are repetitive along the vertical axis...
So, if you used a single ribbon or even a few ribbons stacked (problems with this too...) you would be restricting the vertical dispersion compared to the main array. Maybe this is not the desired result?
_-_-bear
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.