Hi,
I am planning to use LM1875 chipamp to drive the mids of a line array. The LM1875 says 4 ohm is the minimum required load. I am ruling out series/parallel combination of drivers to get 4 ohm since sonically it is inferior to parallel combination. Now, parallel combination yields a very low impedance. So, is it possible to drive impedances lower than 4 ohm if the supply to LM1875 is reduced. The datasheet mentions that it can happily run from a +- 10 volts supply. Since its a line array, I dont expect to draw more than a couple of watts of power, if that helps
Thanks in advance,
Goldy
I am planning to use LM1875 chipamp to drive the mids of a line array. The LM1875 says 4 ohm is the minimum required load. I am ruling out series/parallel combination of drivers to get 4 ohm since sonically it is inferior to parallel combination. Now, parallel combination yields a very low impedance. So, is it possible to drive impedances lower than 4 ohm if the supply to LM1875 is reduced. The datasheet mentions that it can happily run from a +- 10 volts supply. Since its a line array, I dont expect to draw more than a couple of watts of power, if that helps
Thanks in advance,
Goldy
Peter Daniel said:
Hi Peter, As I said its for a line array which would mean lots of amps. I was hoping to get away by using very few amps if it was feasible. Given that only a 4-5 watts (maybe lower coz freq band is 300hz to 3.5khz) of power would be required, is it really possible to lower the impedance to 1 ohm (or 0.5)? Ist'nt the chipamp only bothered about peak current thur it, which can be lowered by reducing the power supply?
AndrewT said:
Hi Andrew, The number of drivers per channel is 24.
Thanks
goldyrathore said:
You need to try it out. I know for a fact that LM3875 may work perfectly fine with 2 ohm loads even when pushed hard, but the heatsinks reach 50 *C or more.
With reduced power and frequency band, 1 ohm shouldn't be a problem either, but it's hard to say what could happen with 0.5 ohms.
Maybe instead of LM1875, 3875 could be more resonbale choice?
Hi,
24 drivers powered by a single 3886 will get <=2W each.
It does not matter whether they are wired up as 2ohm or 3ohm or 6ohm or 8ohm, they still get their share of the 48W maximum that is available.
The amplifier will work better if you load it with a higher impedance load. 6drivers in series and 4sets of 6 in parallel give 6ohm.
24 drivers powered by a single 3886 will get <=2W each.
It does not matter whether they are wired up as 2ohm or 3ohm or 6ohm or 8ohm, they still get their share of the 48W maximum that is available.
The amplifier will work better if you load it with a higher impedance load. 6drivers in series and 4sets of 6 in parallel give 6ohm.
It is a trade-off. The drawbacks of serial connected speakers versus the drawbacks of amplifiers that work into (too) low impedances.
The German magazine Klang&Ton developed a line array two years ago. http://www.lautsprechershop.de/hifi/index.htm?/hifi/twentyfive_en.htm
In that project the drivers are first arranged in parallel groups, and then the groups connected in series. That way the tolerances level out in the parallel connection and the drawbacks of unmatched impedance curves have less impact on the series operation.
The bridged-parallel speaker arrangement gives you the opportunity to try the speakers first with a single amplifier channel per speaker. If the speakers leave you really unsatisfied, don't pursue that project further. If the speakers leave you a little unsatisfied, add more amplifier channels to see, if they improve. If you are satisfied, enjoy the savings.
The German magazine Klang&Ton developed a line array two years ago. http://www.lautsprechershop.de/hifi/index.htm?/hifi/twentyfive_en.htm
In that project the drivers are first arranged in parallel groups, and then the groups connected in series. That way the tolerances level out in the parallel connection and the drawbacks of unmatched impedance curves have less impact on the series operation.
The bridged-parallel speaker arrangement gives you the opportunity to try the speakers first with a single amplifier channel per speaker. If the speakers leave you really unsatisfied, don't pursue that project further. If the speakers leave you a little unsatisfied, add more amplifier channels to see, if they improve. If you are satisfied, enjoy the savings.
Peter Daniel said:
Hi Peter,
Thanks for letting me know that single 1875 may be able drive 1 ohm. I am not sure whteher I can go for bigger chipamps coz that really depends on total no. of amps required. I guess only experimentation can tell.
AndrewT said:
Hi Andrew,
When I reduce the impedance, I also plan to reduce the supply so that the peak current stays within the amps limits. So this way I dont really get 48W and is much lesser than that coz of reduced supply. It is exactly this reduction of supply which, I feel, will let me get away (ok, also the fact that an array only requires a couple of watts) with such a low impedance load.
pacificblue said:
Hi Pacificblue,
Yes, I worked out the configuration you mentioned in the link. Individual drivers do see a very low effective impedance (as compared to 'normal' series/parallel configuration) indeed.
Artie said:Keep in mind, 24 4-ohm drivers in parallel, will be .1667 ohms. One volt will try to push around 6 amps. Not good.
Hi Artie
I am trying to reduce the number of amps depending on whatever impedance the amp can drive comfortably. To enable it to so I feel the supply must be reduced. I dont mean to use only one amp for the entire lot of drivers connected in parallel.
Thank you all for your responses. With this I have gathered a couple of options and will experiment to figure out how each option performs.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.