I'm working on a 3 way speaker design and found a neat small dual coil subwoofer (https://www.daytonaudio.com/product...-mmag-extended-range-subwoofer-4-ohm-per-coil) for the low end speaker. I wanted to try using one coil to cancel out the nonlinearities of the other. To do this, I would use a transformer to act as a comparator, then reverse the polarity to the second coil, thereby acting as an error signal. Is this a feasible way to use a dual coil subwoofer? If so:
1) what kind of transformer should I use?
2) would I need to power both coils, or just treat both of them as part of the same load?
1) what kind of transformer should I use?
2) would I need to power both coils, or just treat both of them as part of the same load?
You should draw a schematic in order to have a basis for reasoning, but I think that if you manage to feed an anti-phase correction signal to the other coil, you will also cancel the useful signal -to be checked-.
Even if it works, the transformer might introduce non-linearities of its own
Even if it works, the transformer might introduce non-linearities of its own
Terrific point. I didn't even think of that. I'm probably going to use a TPA3255 evaluation board for the bass. It has 2 channels and more than enough headroom for that sort of thing. Fourth channel it is!
There are some designs where a dual coil woofer is used, driven by one coil and the other used as error feedback. It is a bit complicated as you have to take into account the transformer that is effectively both coils on the same former. So you need to first determine the transfer curve between the 2 coils when the diaphragm is stationary. Then substract that from the output of the error coil.
Check the article in the Journal of the audio Engineering Society "Transconductance Power Amplifier Systems for Current Driven Loudspeakers" (october 1989) where this type of error correction is described.
Check the article in the Journal of the audio Engineering Society "Transconductance Power Amplifier Systems for Current Driven Loudspeakers" (october 1989) where this type of error correction is described.
Probably not feasible for the Epique E150HE-44, as the primary magnetic drive is provided by the rear ring magnet, directly providing flux to the second rear 7mm gap. The secondary magnet system does not add flux to the upper 7mm gap, but rather acts as magnetic “diodes” (aka gap balancing magnets) pulling the flux from the main magnet into the top 7mm gap.
https://audioxpress.com/article/tes...e-subwoofer-from-the-dayton-audio-epique-line
The driver has good BL (magnetic) and KMS (suspension) symmetry. If you wanted to increase symmetry further, using an external differential pressure sensing device like the Powersoft/B&C IPAL (Integrated Powered Adaptive Loudspeaker) system uses would be a better approach than using half the drive circuit as a comparator/corrector, which would probably generate more nonlinearities than it could correct.
That said, with as linear as the E150HE-44 is when used within it's design limitations, any sonic improvements gained by cancelling nonlinearities are going to be pretty small compared to the expense required to achieve them.
What specific nonlinearities are you finding need correction in the E150HE-44?
So if I am reading this correctly, that would mean that the coils are designed to move in tandem with eachother? In that case yes it would be a bad design choice to do anything that would reduce symmetry between them.
Not a specific one per se, but during my time researching for this little DIY of mine, I came across some literature about bending wave speakers. Between different reading, it reiterated that the speaker cone is always the weakest link in terms of adding any kind of harmonic distortion. I figured if I could smooth out its top end a bit (specifically from about 300hz up where it starts to break up), I could turn my design into a 2 way instead of a 3, cutting down on parts and maybe even making room for something else.
Yes, the coils are designed to move in tandem with each other.
https://audioxpress.com/article/tes...e-subwoofer-from-the-dayton-audio-epique-line
There is no breakup until around 5kHz. The frequency response can easily be corrected with DSP, and the polar response is good to a crossover in the 2500Hz range, easily in two-way territory.
At 94dB at one meter, free-air THD between 100-3kHz is under 1.5%.
You will be using a lot of parts to try to reduce that by any audible amount
.Before you get too far into the distortion reduction rabbit hole, you might try the listening tests here:
https://www.klippel.de/listeningtest/
The audibility threshold for most people that took the test was above the amount of distortion the E150HE-44 has at 94 dB at one meter.
Art
Excellent! I was so focused on it being a sub, but if I can use it as a 2 way then this will definitely make my life easier
Well I guess I have something fun to do after work xD
So @weltersys do you think there is any validity in using the other coil as a reciever of sorts? Or should I just dump the idea altogether and maybe use the second coil for bi-amping? (Since the sensitivity is about 10db less than the other speakes I'm using)
I think that the basic premise of this idea has merit but maybe not with this particular driver. Though short, this has been an interesting read for me. I believe that the idea of using a sensing coil is still in use with such companies as Rhythmic, whom I have purchased a sub from. But as stated above, there was a lot of compensation done their case of a sub, and may not even directly pertain this idea. I just like the fact that you were willing to explore it. Most people want a kit figured out for them.
Using one of the coils as a receiver of sorts will drop the sensitivity and mess up the response.So @weltersys do you think there is any validity in using the other coil as a reciever of sorts? Or should I just dump the idea altogether and maybe use the second coil for bi-amping? (Since the sensitivity is about 10db less than the other speakes I'm using)
"Bi-amping" implies a different frequency range for each coil.
The dual 4 ohm voice coils enable a variety wiring configurations, but not using one or driving either with a different signal would not be a good idea.
Oh I thought it was just a general term for "using more than one amp" to drive a speaker. My bad. Was thinking of using a TPA3255 for that part so I could put one coil on each channel (same signal).
Even though bi-amping is usually taken as using a different amp for separate frequency ranges (typically two different drivers or sets of drivers), it could be understood as you intended to use two channels, one for each coil. I have never tried such a thing but it is not that it cannot be done