Thanks you for the responses. Yes, I am basically locked down to car audio drivers for this application, with the one exception of some Israeli made drivers which the manufacturer has opted not to supply.
I have no intent or need to wire multiple drivers on a single channel but going over the topic of series/parallel wiring again was interesting.
A separate question: I am attempting to source the binding posts on the Morel Soundspot speakers. They appear to be rather conventional brass posts, but they are smaller, more compact. Any suggestions?
Cheers all |
I tend to agree with Duke - the electrical aspect of wiring up two woofers is not such a big deal.
To me the issue is the problem of precise driver matching and the backwave acoustic coupling in the box and the fact that two small woofers with two small voice coils rarely perform like a big bad boy woofer with a massive 4 inch VC. For starters the big VC is going to dissipate heat much more easily due to the larger surface area. The other issue is lobing - by separating the drivers spatially then you run into this problem at lower frequencies than a single driver - this can cause a reduction in reflected energy at certain specific frequencies. |
Duke - I'm leaning toward your view. I think that my previous experience with in-series woofer connection might have to do with other factors. One of them is doubling inductance in the circuit while another is getting stronger resonanses since they might be slightly different, power being proportional to square of the increased voltage on the speaker.
As for "winning" - I have met here many "scientists" with extremely high credentials that prevented them from learning. |
Oops - should be "I have to think about it" - my english is bad but not that bad. |
Kijanki, I gotta tip my hat to you on your openness and honesty in acknowledging that you have found at least one source on the internet that does not agree with your position. Most people would not have posted that, being more interested in "winning".
Best regards.
Duke |
Duke - I like your example wit divided coil, but I think that the difference is that EMF of both coils reacts with the speaker membrane motion while in case of two separate speakers it doesn't - even if they are driven with the same back EMF (same signal). What might happen is that total EMF voltage will be sum of them but impedance each of them sees will be 4 ohm and not sub-ohm amp's output. Two separate speaker coils are just not the same as one coil.
My practical experience tells me that woofers in series have "wooly" sound, but it was long long time ago and my memory is fuzzy. I tried to find something more on the internet to support my point but found instead that opinions are divided. Sweetwater (music store) FAQ agrees with your position - I have to thing about it. |
Oops - that should have been "Kijanki". Sorry I butchered your moniker.
Duke |
Hi Kijanski,
Well, I guess I disagree with JBL! That doesn't seem to put me on very solid ground, does it??
In listing the changes that take place in system characteristics when using two woofers in either series or parallel connection, Vance Dickason (Loudspeaker Design Cookbook) makes no mention of changes in electrical Q nor does he offer any warnings about dropping the amplifier's effective damping factor to less than 1.
I could name several high-end loudspeaker systems that use either series connection or series-parallel connection for two or more woofers. I think that the reason most dual-woofer systems are wired in parallel is in anticipation of the increased wattage output from a voltage-source (solid stae) amp into a lower impedance load.
I have measured loudspeaker systems with series-connected woofers, and my measurements do not indicate that the damping factor goes to 1 with series driver hookup. What I did was this: I measured the same system with a low output impedance solid state amp, and again with a high output impedance specialty tube amp. If the damping factor goes to 1 with series driver connection, there should be negligible change in the bass response between the two amplifier types, as the effective damping factor would be the same in both cases. Instead, I measured approximately 4 dB greater output at system resonance with the high output impedance tube amp (relative to the solid state amp), indicating that the solid state amplifier's high damping factor was in effect. Unfortunately for my position, claimed observations posted in an internet forum are seldom accepted as proof of anything.
Let me present this thought exercise: Think of an 8-ohm woofer's voice coil as consisting of two halves, the first half presenting the amplifier with a 4-ohm load, and the second half also presenting a 4-ohm load, in series with the first. The first half of the voice coil does not isolate the second half of the voice coil from the amplifier. Instead, the amplifier sees them together as a single 8-ohm coil.
Now suppose that, instead of those two halves of the voice coil being continuous, we connect them with a short piece of wire that comes off the voice coil former, and then returns to it. Would this make any significant difference in the load that the amplifier sees? No, neither the impedance nor the inductance of the load is changed.
So let's suppose that we use a separate voice coil former for the second half of that voice coil, along with a separate motor and frame and spider and cone (in other words, we now move to two 4-ohm drivers connected in series). Would that make any significant difference in the load that the amplifier sees? Again, the answer is no.
Duke |
System doesn't like long links and shortens them.
The way I see it speakers is controlled when EMF generated by the speaker is
shorted. Just take speaker and move/press its membrane - it moves freely.
Now try to repeat it with speaker shorted by wire - membrane won't move.
When you have two speakers in series one sees impedance of another in
series. No matter what will happen with EMFs of two identical speakers
impedance of another speaker won't disappear.
DF is defined as speaker (and not the speaker box) impedance over amps
impedance. If my 8 ohm woofer has 0.1 ohm inductor in series inside of the
box DF will drop to 80 even if amplifier's output impedance is zero.
As far as I know speaker with identical woofers have them always in- parallel.
There are even bass players forums where they mention poor definition of in
series wired speakers. |
Sorry - link got shortened in my last post - here it is again:
http://manuals.harman.com/JBL/CAR/Owner's%20Manual/15515%20GT4%20Man_ENG.pdf |
Audiokinesis - JBL in users manual for GT4 car subwoofer system states:
"We recommend that you avoid
connecting separate woofers in
series. The amplifier-damping
factor (the amplifiers ability to
control the motion of the woofer
is expressed as a ratio of termin
impedance (the sum of speaker
impedance, wire resistance and
the D.C. resistance of any
crossover coil connected to the
woofer) to amplifier-output
impedance. Therefore, connectin
separate woofers in series
reduces the damping factor of
the amplifier to a value less
than 1. This will result in poor
transient response. "
and the link is here: http://manuals.harman.com/JBL/CAR/Owner's%20Manual/15515%20GT4%20Man_ENG.pdf
Of course they might be wrong. |
It looks like my previous answer didn't make it past the moderators. I have no idea why.
Connecting two drivers in series does not lower the amplifier's effective damping factor - instead, it raises it. Damping factor is the ratio of speaker input impedance to amplifier output impedance.
I have some experience working with series-wired drivers.
Duke
dealer/manufacturer |
When I was fooling around with adding stereo speakers to my 57 Packard in the "good ol' days" I seem to remember that some speakers had transformers attached to them. ( Sweet ride..staight 8, used more oil than gas) |
...Do not connect speakers in series... Point taken, but this is a "pick your poison" issue. I'd take possibly making my bass quality a bit worse off as opposed to burning out an amp. You will recall that our original poster stated that "only a certain high-end car speakers fit the design criteria." It sounds like he has looked for other drivers and not found any. In situations like that, one is often left with choosing the lesser of two evils. |
"The only practical way to increase driver impedance is to either wire two drivers in series or use a transformer."
Do not connect speakers in series because it will lower damping by amplifier (speakers isolate each other from the amp) - poor bass definition. |
>> there is concern about their low impedance damaging the
>> amplifier. Driver impedance ranges from 2.8 ohms to 3.4
>> ohms. Will this do harm?
it depends on the model of the amplifier. To get a hint of its performance look @ the amp's o/p specs. Does the output power double each time the load resistance halves? if it does,it would indicate an amp with a fairly robust power supply & the chances that such an amp could handle a lower speaker impedance are reasonably high.
if the amp power does not double the amp could still drive this lower speaker impedance but the amp will run out of juice. This'll manifest itself as lacklustre reproduced sound. If you crank up the volume, the amp could clip resulting in a distorted sound.
>> And if several speakers are used in a 5.1 surround
>> setup, does this increase the load and the chance of
>> damage?
should I read this to mean that you are going to connect 5 speakers to 1 pair of binding posts????
If yes, I'd say that's totally assinine!!
If no, then you must be using a 5 channel amp that has 1 pair of binding posts for each speaker + an output for the sub. In that case, you have 5 independent amplifiers each driving just 1 speaker so the question of overload does not occur if you keep the volume control within the limits of these amps per the amp spec sheet. Yes, there are all getting power from 1 transformer so each is "connected" to all the other amps & all 5 amps are not truely independent but this is a 2nd order effect for this discussion
>> If the low impedance is indeed risky, is their any
>> means of raising it perhaps by placing something in the
>> chain
you can create a Zobel network which'll "fool" the amp that it's seeing a (more) constant load vs. freq but I would advise against attaching such a network if you are not a serious DIYer or if you do not know much about speaker design or if the speaker was never meant to be used w/ a Zobel network. You will surely alter the sound (more often for the worse but you might get lucky & change it for the better! who knows...). Of course contact the speaker manuf for any advice they can offer re. Zobel networks for your speaker OR see if there is any discussion on your speaker model in some audio forum.
The other solutions are to either get a better amp to handle the lower speaker impedance OR get another speaker that is an easier load to your amplifier.
FWIW. |
It completely depends on the amplifier involved. Some amps do well with low impedance loads, others don't.
Among the amps that don't do well with low impedance loads the results can vary from distortion or the protection circuit engaging all the way to damaging the amp by shorting the output devices.
Check your amp's owners manual.
The only practical way to increase driver impedance is to either wire two drivers in series or use a transformer. (Transformers are normally used on tube amps for impedance matching but they can be used on solid state amps, too.) |
