Ok to wire line array in series? Why?

[jaekfre]

I can't find anything info dealing with this on diyaudio or any place else.

I've been told and read in many places that one should not wire two drivers in series because of the back EMF that it produces.

However, why is that one of the accepted ways of wiring line arrays?

Makes no sense to me. But of course, I'm new to speaker building.

Thanks in advance.

 

[chris661]

Most people go for series/parallel groups that end up at 4-8ohm. This means the load on the amplifier is what it's designed for. If you had 9 8ohm speakers in series, you would end up at 72ohm, which the amp will drive, just very quietly.

So, to take the 9x 8ohm speakers, if you made 3 groups of parallels with 3 drivers each, you'd get an 8ohm load exactly (neat, huh?) and your amp would be fine with that.

If, however, you wired all the 9 speakers in parallel, you'd get a 0.9ohm load, which the amp wouldn't go well with. It would **try** to give 3x it's rated power (into 8ohm) into the speaker, which would lead to damage and nasty smells.

Basic rules (dunno if you know this, handy to have) calculating overall impedance:

two speakers in series: add the impedances together.
two speakers in parallel: (gets more complicated) you do 1/(1/R+1/R+1/R.....) where R is the impedance, and the number of 1/R s is how many speakers there is.

Hope this helps

Chris

PS - you shouldn't wire two *different* drivers in series, as bad things can happen (according to Dad, the electrical engineer, off whom I got the info for the impedance calculations).

 

[slam]

I once opened up a McInstosh XRT-22 line array to fix a loose connection and found that the 23 tweeters were in a mixed parallel / series configuration to get to around 8 ohms.

BTW those speakers sound very good. So don't worry.

 

[jaekfre]

That makes sense.

But from what I've read is that no two drivers are identical - even though they be the 'same' driver from the same manufacture.

Supposedly, the slight variations will cause distortion because of back EMF.

Is this something that is theoretically true but not going to be a huge deal with just two of the 'same' drivers wired in series?

 

[jaekfre]

ok, i'm still in 'moderation mode' until i make some more posts so the lag in my posts showing up is pretty annoying!

anyway, thanks a lot guys for the replies.

guess i'll try some experimenting

 

[jaekfre]

That makes sense.

But from what I've read is that no two drivers are identical - even though they be the 'same' driver from the same manufacture.

Supposedly, the slight variations will cause distortion because of back EMF.

Is this something that is theoretically true but not going to be a huge deal with just two of the 'same' drivers wired in series?

Slam - I posted the above before you replied to the thread.

You pretty much answered the question. Thanks again.

 

[Cal Weldon]

You are now registered.

 

[kipman725]

yep series parallel works listening to 9 drivers sereis paralel right now:
+ 3 > 3 > 3 - It's just not sane to try and drive loads of less than an ohm!

 

[jaekfre]

awesome! thanks kipman.

 

[dartamaji kachapury]

thank,s for all

 

[flg]

Interesting, I've missed this thread, years ago? But, never the less, I have heard the same claims without explaination? The parallel connection lowers inductance and DCR but deserves a very good amp if you want performance. A benifit may also be that you do not need as much voltage gain in such a system.
Personally, I don't think we got an answer to the original poster's questions other than "it's OK" to wire it that way?
I think we now have an expert Line array guy here. Speaker Dave, on this forum may have actually been directly involved in the XRT mentioned above and obviously one of the most accomplished Line Array guy's we can easily access for such answers 😀

 

[wesayso]

Maybe an answer from Roger Russell will do:

The characteristics of the drivers I used are literally identical and the impedance curve of the combined drivers matches one individual driver. Perhaps in the old days manufacturing control was not that tight but there is no problem with the drivers I have used.

Someone has suggested all the drivers must be in parallel. Again, this is more of an opinion than fact and in addition introduces an impedance matching problem.

I got this answer from Roger Russell some years ago when asking almost the same question.

I believe he was involved in the development of the XRT-22 😉: McIntosh XRT22 Speaker System

 

[flg]

If Speaker Dave didn't work with R. Russell, he took over his work at McIntosh when Roger left, IIRC. Dave is also a veteran of Snell Acoustics... He has a few papers floating around worth reading also... Look him up 😉

 

[wesayso]

I know the papers you talk off, I've read them. But Roger should be a reasonable source in this case, when talking about the ideas behind the XRT-22.
The quote I posted was about his IDS-25 speaker though, I should mention that!

 

[Deep Throat]

Hi impedance question!
Since it is the current which effectuates the motion in an electrodynamic motor of a driver, best to ensure same motion for all drivers is wiring in series.
Tolerance of drivers lead to impairment in both wiring schemes to the same degree, hence in different areas.
Often overseen is the non-identical acoustic loading of the drivers (i.e.: one woofer closer to the floor, and "seeing" the room at another point, which then causes via impedance modulation different motion with parallel wiring, for such effects the series one is immune of.
Unfortunately are very low impedance drivers rare, but there are big manufacturers which do it with these in series.
Deep Throat

 

[fpitas]

Series is fine if the drivers are close to identical:

http://www.cartchunk.org/audiotopics/SeriesSpeaker.pdf

 

[PeteMcK]

back EMF shouldn't be a problem for home audio, maybe a problem for high power situations, e.g. guitar quad boxes, which is why they should be wired as two parallel pairs in series rather than two series pairs in parallel...