Daft question of the day, sorry !
If I have chosen a woofer speaker that has both 8 and 4 ohms (nominal) versions (similar Fs, Qts, Vas so can use similar boxes) and wanted to run two for bass duties in a 3 way -
8ohm - sensitivity - 86dB 1W/1m
4ohm - sensitivity - 87.5dB 1W/1m
is is best to run two 8ohm in parallel (nominal 4ohm) or two 4 ohm in series (8ohm nominal) understanding that the 8ohm will have higher combined sensitivity but will be offset by the 4ohm higher intrinsic sensitivity?
Assume the amp will run down to 2ohm (just) and has plenty of power (250w into 4 ohm) and the crossover will be a dsp so can match the mid/tweeter easily.
If I have chosen a woofer speaker that has both 8 and 4 ohms (nominal) versions (similar Fs, Qts, Vas so can use similar boxes) and wanted to run two for bass duties in a 3 way -
8ohm - sensitivity - 86dB 1W/1m
4ohm - sensitivity - 87.5dB 1W/1m
is is best to run two 8ohm in parallel (nominal 4ohm) or two 4 ohm in series (8ohm nominal) understanding that the 8ohm will have higher combined sensitivity but will be offset by the 4ohm higher intrinsic sensitivity?
Assume the amp will run down to 2ohm (just) and has plenty of power (250w into 4 ohm) and the crossover will be a dsp so can match the mid/tweeter easily.
I would prefer the parallel option. Mainly because if any of two results disconnected, it will continue sounding, but more important is that if the amp behaves like a voltage source, their effect between them will be almost cancelled electrically. Search in any acoustic book that the equivalent electric of the speakers are almost all elements in series (elastances, inertances and resistances), then wiring them in series will result in a serious interaction between their independent behavior.
When deciding on this kind of thing and doing a passive XO, then i look at the net sensitivity vrs what is needed to match the other driver(s). For instance in Tysen V2 we use 2 4Ω in series to get a sensitivity that means no Rs in the midTweeter XO. If i was bi-amping i’d use 2 8Ω in parallel.
dave
dave
Series or parallel?
It is often claimed that putting speakers in series is a bad idea because the series resistance of one speaker destroys the damping of the other.
However analysis shows that the damping is not severely compromised by connecting two drivers in series, because the total Q remains the same.
You may wish to consider opting for the series, 8 ohm connection to provide a non-challenging amplifier load.
It is often claimed that putting speakers in series is a bad idea because the series resistance of one speaker destroys the damping of the other.
However analysis shows that the damping is not severely compromised by connecting two drivers in series, because the total Q remains the same.
You may wish to consider opting for the series, 8 ohm connection to provide a non-challenging amplifier load.
I say this purely theoretically, but Lynn Olson mentioned his reason for putting two parallel wired 5" Vifa P13WH drivers in seperate enclosures in his famous Ariel MTM speaker.
It was to stop driver interaction. If the two drivers are in a common (closed box) enclosure, push one in, the other pops out. Something similar must go on with reflex, particularly at the resonant Fs impedance point where impedance goes very high, and possibly slightly mismatched.
You can take this or leave it, but my theoretical preference would be two 4 ohm (and more efficient) units wired in series for the 8 ohms amps prefer.
No problems with damping factor, which just all falls into place, just double the common enclosure volume of a single driver and double the reflex tubes. The way I see it, both woofers get the same current when series wired. So move together. This idea gets little love with the parallel MTM community because it is the same loudness as a single driver. But it is at half the acoustic power in the room, so has the nice PA directivity and sound.
Nice little thought experiment below. So prove me wrong!
It was to stop driver interaction. If the two drivers are in a common (closed box) enclosure, push one in, the other pops out. Something similar must go on with reflex, particularly at the resonant Fs impedance point where impedance goes very high, and possibly slightly mismatched.
You can take this or leave it, but my theoretical preference would be two 4 ohm (and more efficient) units wired in series for the 8 ohms amps prefer.
No problems with damping factor, which just all falls into place, just double the common enclosure volume of a single driver and double the reflex tubes. The way I see it, both woofers get the same current when series wired. So move together. This idea gets little love with the parallel MTM community because it is the same loudness as a single driver. But it is at half the acoustic power in the room, so has the nice PA directivity and sound.
Nice little thought experiment below. So prove me wrong!
Attachments
My theoretical and practical preference would be two 8 ohm units wired in parallel, with total 4-ohm impedance. No problem with OP amplifier which is stable in 4 and 2 ohms.
Sensitivity of two 4-ohm units wired in series will be 87.5 dB/2.83V (the same as one 4-ohm unit).
Sensitivity of two 8-ohm units wired in wired will be 92 dB/2.83V (6 dB more than one 8-ohm unit). Contrary to the naive opinion, both units will get the same current, because they have identical impedance. Simple Ohm's law. Also, they have the same PA directivity as serial connected MTM.
The same current can mean different voltages across the 2 drivers unless there impedance curves are identical. And since the drivers are designed to be voltage driven you will end up with non flat freq. response of the drivers. Say one driver has a impedance peak at 100hz the other at 120hz. With a 100hz signal the 120hz driver will be producing most of the sound. So be careful using 2 different types of drivers in series.
Many thanks guys, much appreciated.
I think I should have made this a vote
The drivers I am looking at are the diysoundgroup Anarchy which I know the 8 ohm are out of stock but fingers crossed for another run.
Looks like these (both 4 and 8 ohm) work well in a 20l box with f3 in the low 30’s so could go with two in a 40l slim floor standers or separate the box into two 20l compartments. Looking to marry these with the morel em1038 and sb26adc or cdc with crossovers at around 800hz and 4khz. This models quite well in boxsim with 4th order slopes and the odd notch filter to smooth it out.
Keep the comments coming as I am still undecided.
I think I should have made this a vote
The drivers I am looking at are the diysoundgroup Anarchy which I know the 8 ohm are out of stock but fingers crossed for another run.
Looks like these (both 4 and 8 ohm) work well in a 20l box with f3 in the low 30’s so could go with two in a 40l slim floor standers or separate the box into two 20l compartments. Looking to marry these with the morel em1038 and sb26adc or cdc with crossovers at around 800hz and 4khz. This models quite well in boxsim with 4th order slopes and the odd notch filter to smooth it out.
Keep the comments coming as I am still undecided.
I got in terrible trouble with Craig at Selah Audio for posting a picture of his splendid MTM ribbon product without a credit. TBH, I think he overreacted.
But here's why we like MTM with ribbon. It's a lot of trouble and expense to do MTM, but has the splendid advantage of reducing power for the same loudness on vertical axis.
Which delivers more loudness to you in your apartment, whilst annoying your neighbours less. Frankly, a dome tweeter would work less well. A question of dispersion and the inverse square law of physics.
The more you move towards line array, and MTM ribbon is the first sensible step, the less you upset the neighbours. We proved this at a recent disco at a workplace. Line array sounded less annoying outside the venue when the windows were open.
But here's why we like MTM with ribbon. It's a lot of trouble and expense to do MTM, but has the splendid advantage of reducing power for the same loudness on vertical axis.
Which delivers more loudness to you in your apartment, whilst annoying your neighbours less. Frankly, a dome tweeter would work less well. A question of dispersion and the inverse square law of physics.
The more you move towards line array, and MTM ribbon is the first sensible step, the less you upset the neighbours. We proved this at a recent disco at a workplace. Line array sounded less annoying outside the venue when the windows were open.
Attachments
As an observation, the reason Lynn put the midbass drivers in isolated pipes / lines in the Ariel from V4 onward was this: in the V2 and V3 versions of the Ariel enclosure, two lines of unequal length were coupled together when they both reached what Lynn describes as the labyrinth, by which he is referring to the highly folded section that forms the bottom of the line. As a result, he got some unequal acoustical loading and as this was not optimised (differential tuning becomes harder when two lines are physically combined into one) resulted in some smearing in the lower midband. His solution was to isolate the drivers in independent lines of equal length, from the V4 version onward. This works and solved the issue. As it happens, you can also prevent this issue by simply ensuring the drivers share the same line without having separate branches of unequal length. Providing the drivers are acoustically close together, which is not difficult given the wavelengths in question, the pipe they load 'sees' a single drive unit, with the centre of excitation being the acoustic centre point between the drivers.
Last edited:
Ah, but this is where Ohm's law gets a little abused.
When specs state a 4 ohm driver is 87.5 db 1W/1M, are they basing this on 2 volts (the law) or 2.83 volts? If 2 volts, it's 2.83 volt sensitivity would be 90.5 dB at 1 meter, I believe.
Yes, you are right, their sensitivity is specified as 1W/1m, not 2.83V/1m, my bad.
So, sensitivity of two 4-ohm units wired in series will be 90.5 dB/2.83V (the same as one 4-ohm unit).
Sensitivity of two 8-ohm units wired in parallel will be 92 dB/2.83V (6 dB more than one 8-ohm unit).
Still, parallel wiring brings 2 dB more with the same voltage input. And is not sensitive to unit-to-unit variation in impedance, contrary to the serial wiring.
So, sensitivity of two 4-ohm units wired in series will be 90.5 dB/2.83V (the same as one 4-ohm unit).
Sensitivity of two 8-ohm units wired in parallel will be 92 dB/2.83V (6 dB more than one 8-ohm unit).
Still, parallel wiring brings 2 dB more with the same voltage input. And is not sensitive to unit-to-unit variation in impedance, contrary to the serial wiring.
I have often found a driver specced at 1w/1m but they are actually 2.83 v/1m.
All the same rukes apply, but the starting numbers can sometimes be misleading.
If you take 2 XΩ drivers then one driver will have the same sensitivity of 2 in series, but efficiency is up 3 dB (ie 3 dB from double the number of drivers, -3dB as the impedance doubles.
If you take 2 XΩ drivers in parallel, you gain 3dB from the doubling and 3 dB from the halving of the impedance.
dave
All the same rukes apply, but the starting numbers can sometimes be misleading.
If you take 2 XΩ drivers then one driver will have the same sensitivity of 2 in series, but efficiency is up 3 dB (ie 3 dB from double the number of drivers, -3dB as the impedance doubles.
If you take 2 XΩ drivers in parallel, you gain 3dB from the doubling and 3 dB from the halving of the impedance.
dave
I got in terrible trouble with Craig at Selah Audio for posting a picture of his splendid MTM ribbon product without a credit. TBH, I think he overreacted.
But here's why we like MTM with ribbon. It's a lot of trouble and expense to do MTM, but has the splendid advantage of reducing power for the same loudness on vertical axis.
Which delivers more loudness to you in your apartment, whilst annoying your neighbours less. Frankly, a dome tweeter would work less well. A question of dispersion and the inverse square law of physics.
The more you move towards line array, and MTM ribbon is the first sensible step, the less you upset the neighbours. We proved this at a recent disco at a workplace. Line array sounded less annoying outside the venue when the windows were open.
It wasn't terrible trouble - I simply asked.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.