Are two separate lm3886 = 2*lm3886 in parallel or bridge mode

[rs1026]

I had a basic doubt. If I drive 2 drivers capable of delivering 60W each, separately with 2 separate lm3886 based amps then is it equal to driving one driver delivering 120W using 2 lm3886 in parallel or bridge mode?

I hope my question is clear

[Ed LaFontaine]

The National Semi application note found here: http://www.national.com/an/AN/AN-1192.pdf provides examples of bridge and parallel solutions.
Generally speaking,
In a bridged amp each LM3886 would see 1/2 the impedance. The bridged amp would consume 4 times the power AND have to disipate the excess as heat. Better heatsinking is required.

In a parallel amp each LM3886 would see the total impedance. The parallel amp would provide twice the drive current. The chips in a parallel amp share the heat load, so less heatsinking is required.

Each of the above has +'s and -'s. Relative to one amp, one chip

It is all in the app note-and more! Take a look.

[rs1026]

Thanks bg, I am clear about bridged and parallel mode conneions.But what I wanted to know is if I connect 2 drivers in same box using 2 diff lm3886 chips then the sum total of power equals say roughly 100W is this power euquivalent to 100 watts obtained by bridged or parallel connection.

thanks

[Ed LaFontaine]

rs1026,
100W=100W under the same test conditions, but with the same power supply you would get different watts from bridge, parallel or discrete amps. It depends on the amp configuration.
Using a different amp configuration changes the test condition, so you would obtain a different wattage.

On the other side you can see that parallel speakers would have half the resistance, therefore the wattage increases (maybe doubles) to the ability of the amp.

I hope this helps

[sam9]

In bridge mode where each one sees 1/2 full speaker impedance, you will get increased disdortion. In general with BJT's lower impedance means higer distortion. Going from 8 ohms to 4 ohms can often double the THD figure. Whether you feel this is a problem or not depends on what you think a reasonable THD figure is.

In parrallel mode where the impedance is not affected, this is a non-issue.

I can't say one is better or not, but you should be aware of it when making your own decision.

[troystg]

Hi rs-

"2 drivers in same box using 2 diff lm3886 chips " = ~ 60 watts x 2 channels.

60 watts per driver. Sum power from box to drivers =~ 120 watts.

Parallel or bridge the 2 LM3886 chips and get either more current or more voltage respectfully.

[sam9]

Check the loudspeaker forum regarding two drivers in the same box. There may be some issues regarding lobing, imaging etc. Speaker design is much more esoteric and obscure than mere amplifiers!

[carlosfm]

I understand RS1026's question.
His doubt is, for instance, if an amp with five channels, 100w each can be considered a 500w amp.
No.
It's a 5x100w amp.
Less serious manufacturers would put a sticker saying something like "500 watts amp", or even worse, 10,000 watts PMPO.

RS1026's doubt is if it's better to bi-amp or to single-amp with double the power.
It depends on the amp and the speaker.
If you were going to buy commercial products you could test one way or the other and decide what works best with a particular speaker.
In diy you make it and test it.
Anyway, talking in paralleling chips (doublig the current, usually more important than getting more power), it's easy to make a 4-channel amp (with single RCA input) and test it in bi-amp mode or parallel.
Use resistors in series with each chip's output, as recommended, and then to parallel is just a question of inserting a wire link between the speaker's binding posts on the amp.
Easy.

[rs1026]

Thanks guys, carlos got my question right
Other inputs were also helpful.

[tuan_sheikh]

dear all

If I have an 8 ohm speaker (let assume a true 8 ohm) it is better for me to have 2 chip in parallel config rather then to bridge ?.

Reason = double the current but remain the power ?.

Thanks