Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

how to drive a 0.4 ohms load?

Status
Not open for further replies.
Actually, the OP wants 56V p-p to each 4 ohm speaker which is 100W that is too much for one LM3886 amp. The applications note shows two amps in parallel (with low value series resistors at their outputs) then the power into 4 ohms is about 110W.

As I told I have taken apart each speaker from an amplifier module containing one LM3886. I want they work in my liner speaker design just like they have done before (i.e while they were connected to said module), Again, I will not get such situation if I put whole speakers in series/parallel.

By put all speakers in parallel the net impedance across the net connection would be 0.4 ohms, It means that using a bridged amplifier out of 2 LM3886's would not be helpful (by using a bridged amplifier out of two LM3886's I can drive a 8 ohms load which is completely useless).
There are 2 methods in my mind (I do not know any reasonable third method yet having the maximum output power for each speaker if needed).
1.feeding each speaker through separate amplifiers.
2.feeding whole speakers connected in parallel (thous having a 0.4 ohms load) through a Parallel amplifier (i.e using several LM3886's and put the output of them all in parallel).

Which method do you think is better please?

P.S Can I use 2 or more power transistors as drivers somehow to honestly be able to connect the 0.4 ohms loads to them without any problem yet getting the maximum of power??
 
I still don't understand why you're so dead-set against simply wiring your speaker array in series-parallel.

OK, forget speakers. Let's talk light bulbs instead. Let's say you're working for some crazy guy who wants you to wire up 16 100-watt lights bulbs to that they all burn at maximum brightness. (This is in the country of Freedonia, where the standard household current is 100 volts, for simplicity's sake.) But for some strange reason, he insists that you cannot just wire them in parallel like you'd normally do.

So you come up with the following plan:

**broken link removed**

Of course, the crazy guy has also supplied you with the 400 volts necessary to run four bulbs in series.

The point is, do you see that each bulb in this array burns with the proper voltage and current, and gets its maximum allowable power (100 watts)?

So series-parallel will work.

(By the way, to the O.P., you do mean "linear" speaker system, correct? I still don't know what that is; guess I'll have to look it up.)
 
Last edited:
I believe we are having a problem in semantics. Exactly what you mean by getting the "maximum power out of each"?

Maybe the problem is caused due to my broken English?!

If the amp is putting out it's maximum power than that is all it can do. With proper series-parallel connection of the speakers, they will have the combined equivalent impedance for which the amplifier delivers its maximum power. But of course that power will divided by the number of speakers. What more could you expect? You can't put more total power into the speakers than an amp can deliver. For more power you need to go to more amplifiers.

By saying "maximum of power" I just mean that I want each speaker while used in my liner speaker design (i.e used together with other speakers, for instance by paralleling the speakers or use separate chips for each speaker or using any other possible method) works just like while used by each amplifier while I bought it as a module (every speaker originally was connected to a circuit containing one LM3886).
Hope it makes sense for now. I do not know how to say this the other way!!

I already know that by putting the speakers in series then in parallel I can get the desired impedance (i.e 4 ohms in this case which is necessary to not cause the single LM3886 to burns out), But I just trying to say That I will not get the maximum of power out of each speaker to have a louder output.
 
I still don't understand why you're so dead-set against simply wiring your speaker array in series-parallel.

OK, forget speakers. Let's talk light bulbs instead. Let's say you're working for some crazy guy who wants you to wire up 16 100-watt lights bulbs to that they all burn at maximum brightness. (This is in the country of Freedonia, where the standard household current is 100 volts, for simplicity's sake.) But for some strange reason, he insists that you cannot just wire them in parallel like you'd normally do.

So you come up with the following plan:

**broken link removed**

Of course, the crazy guy has also supplied you with the 400 volts necessary to run four bulbs in series.

The point is, do you see that each bulb in this array burns with the proper voltage and current, and gets its maximum allowable power (100 watts)?

So series-parallel will work.

(By the way, to the O.P., you do mean "linear" speaker system, correct? I still don't know what that is; guess I'll have to look it up.)

Thanks for your efforts, but maybe it is better I go and kill myself:rolleyes:!!!

In your case, every bulb is Arbitrary chosen as 100V and the net source voltage is 400V.
Now please tell me if you could make those 4 bulbs in series if they were labeled as 400V each, yet supplying them by the 400V source and get the MAXIMUM of power out of each bulb???!!....

Suppose you have a bulb which is satisfied to work by mains (i.e 110V), Now are you able to use 2 of the said bulbs is series yet connecting them to your mains or you have to increase the main voltage by factor 2 to light up each bulb at its max or say normal power?
 
Last edited:
Take Carl's suggestion of using an IC for each speaker. The delay (or phase shift) through the IC will not be a problem. You will need one big power supply!
 
Take Carl's suggestion of using an IC for each speaker. The delay (or phase shift) through the IC will not be a problem. You will need one big power supply!

I second that emotion.

Concerning the phase problem, whatever phase shift there is should be nearly identical from amplifier to amplfier, no? So not a problem for the ultimate resulting combined audio wave from all speakers.
 
It is obvious that for an output of 68W into each 4 ohm speaker then each speaker needs its own LM3886 amplifier powered from a 56VDC or plus and minus 28VDC supply. Forget about series-parallel or a transformer.
With a 60VDC supply then the total of ten amplifiers is 850W which is too high for them.
For a continuous output of 680W then the power transformer must be huge and rated for about 1244VA.
 
So the consensus here seems to be converging towards putting a separate power amplifier on each and ever speaker, with a massive ginormous transformer powering them all.

Is what you're trying to do actually done in the Real World? If so, how do they power their speakers? Hard to imagine them doing things the way you're proposing to do it; it seems like your project is being driven by your choice of what stuff you have to work with, which is kind of bass-ackwards from an engineering standpoint.

I'm guessing the ultimate application is towers of speakers to blast super-loud rock music towards audiences, who will eventually notice an acute loss of upper-frequency hearing ability when they hit their 40s or 50s. Am I warm?
 
Last edited:
About 37 years ago I measured a sound pressure level of 120dB in a discotech. It was the low frequency beat that was loud, not all the music and the beat was for only a moment every half-second. I could feel the beat and hear it but it wasn't too loud. If the sound was distorted then it would sound too loud and damage your hearing.

I Assembled a demo system in the Canadian head office of a Japanese Pro-Sound manufacturer. The two 15" sub-woofers were very sensitive and were driven with 1800W RMS. The stereo speakers had 8" woofers and compression horn tweeters and were also very sensitive and were each fed 220W RMS. The system was loud but not deafening since I never fed it with continuous sine-waves or acid rock.

Continuous wide-range sound at 90dB for a few hours will make you very deaf. Continuous wide-range sound at 80dB for 8 hours for many days will make you deaf to the high frequencies of music but you still might understand telephone speech so you will not be called deaf.
 
Take Carl's suggestion of using an IC for each speaker. The delay (or phase shift) through the IC will not be a problem. You will need one big power supply!

Ok, it seems that I have to do so, even though I did not get why not use a Parallel configuration of LM3886 chips?! Or if it is possible using a push-pull very powerful configuration on the output of one LM3886 to get higher power to drive a very low load?

Yea, Now I have a problem regarding the power supply for the device. suppose I decide to use a 56 to 60V center tapped supply (+-28 to +-30V), then how many current the transformer should be able to deliver to the said 0.4 ohms load (i.e parallel speakers) without any problem?? is a 12A, transformer capable of delivering 60V DC (+-30V) enough for the job?
 
It is obvious that for an output of 68W into each 4 ohm speaker then each speaker needs its own LM3886 amplifier powered from a 56VDC or plus and minus 28VDC supply. Forget about series-parallel or a transformer.QUOTE]

Yea, Thats right.

With a 60VDC supply then the total of ten amplifiers is 850W which is too high for them.
For a continuous output of 680W then the power transformer must be huge and rated for about 1244VA.

Sorry but What "1244VA" does mean? Is it Volt-Ampere or VAR? how to interpret it for a 60V transformer? Is it voltage x current?....
 
Sorry but What "1244VA" does mean? Is it Volt-Ampere or VAR? how to interpret it for a 60V transformer? Is it voltage x current?....
Yes, VA is volt-amps. (VAR is volt-amps reactive). Transformers are rated in VA because they are limited by the current they can deliver (due to internal I²R heating from the winding resistance), not the power output. For example if the load was totally reactive you could have high current from the transformer but no power. If the load is pure resistive than volt-amps out equals watts.
 
Don't you know that a 60V transformer produces rectified and filtered 83VDC, not 60VDC. 83VDC is much too high for an LM3886 that drives 4 ohms. You need a 41V transformer which will produce rectified and filtered 56VDC. It must have a current rating of 22.2A if the amplifiers blast 68W each continuously.
 
Don't you know that a 60V transformer produces rectified and filtered 83VDC, not 60VDC. 83VDC is much too high for an LM3886 that drives 4 ohms. You need a 41V transformer which will produce rectified and filtered 56VDC. It must have a current rating of 22.2A if the amplifiers blast 68W each continuously.

Of course I now that a 60V AC transformer will produce 60x1.4 DC value. Actually I now that it is the RMS value, thats why I told that the output of the transformer is 60v DC, which means almost 42.5V RMS....


22.2A?But Why? I was thinking of 13 to 14A...
 
22.2A?But Why? I was thinking of 13 to 14A...
In addition to the 680W to the speakers, the amplifiers heat with 560W. The transformer must also supply this power.

You can cheat a little with the power from the transformer because your program is music or speech, not continuous sine-waves.
 
Question for the experts: amplifiers in parallel

I was going to chide "Dr. Power", who seems hell-bent on putting all his speakers in parallel and burning out his amplifiers, until it occurred to me: I don't really know what happens when one puts two or more amplifiers in parallel with each other.

This ASS-U-MEs, of course, that the amplifiers are designed for this usage (bridge-able, I guess you could say). And to reduce confusion in this discussion, let's limit this to transistor amplifiers without output transformers (in other words, 99% of amps found today).

What happens to the total impedance when amplifiers are put in parallel? Both from the point of view of the amplifiers and their loads (i.e., loudspeakers). Do the impedances reduce the same way resistances in parallel do?

In other words, could "Dr. Power" actually get away with, say, putting 10 LM3886s in parallel and running 10 4-ohm speakers also in parallel, for a total load of 0.4Ω, as per the title of this thread?

Or is he just going to produce a lot of expensive smoke this way? (My money would be on this outcome.)

Inquiring minds want to know ...
 
I was going to chide "Dr. Power", who seems hell-bent on putting all his speakers in parallel and burning out his amplifiers, until it occurred to me: I don't really know what happens when one puts two or more amplifiers in parallel with each other.

This ASS-U-MEs, of course, that the amplifiers are designed for this usage (bridge-able, I guess you could say). And to reduce confusion in this discussion, let's limit this to transistor amplifiers without output transformers (in other words, 99% of amps found today).

What happens to the total impedance when amplifiers are put in parallel? Both from the point of view of the amplifiers and their loads (i.e., loudspeakers). Do the impedances reduce the same way resistances in parallel do?

In other words, could "Dr. Power" actually get away with, say, putting 10 LM3886s in parallel and running 10 4-ohm speakers also in parallel, for a total load of 0.4Ω, as per the title of this thread?

Or is he just going to produce a lot of expensive smoke this way? (My money would be on this outcome.)

Inquiring minds want to know ...

To a certain extent, and with correctly designed circuits, you 'can' parallel identical amps.

However - what would be the point?. If you've got ten speakers, and ten amps, then simply feed each speaker from it's own amp - which will be at least (and probably more) ten times as reliable.
 
To a certain extent, and with correctly designed circuits, you 'can' parallel identical amps.

However - what would be the point?. If you've got ten speakers, and ten amps, then simply feed each speaker from it's own amp - which will be at least (and probably more) ten times as reliable.

Yes, of course, which is what lots of folks seem to have been suggesting here, with the suggestion falling on deaf ears ...

Mine is more of a pedagogical point (IOW, I'd like to learn something from this): what happens to the impedances of amplifiers in parallel?

As you said, you can (under certain conditions) parallel amps; but I think the O.P. is under the mistaken impression (at least I think it's mistaken) that amplifier impedances behave much the same as, say, resistances in parallel. The thinking seems to be "If I have 10 speakers in parallel (a load of 0.4Ω). then hey, I'll just put 10 amplifiers in parallel, with a total impedance of 0.4Ω, so they should match."

To try to answer my own question, I guess you would, what? try to Thévenize the amplifier/speaker network? reducing the amplifiers to their internal resistances and current/voltage sources? although how in the world you would do that, I have no idea.
 
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top