• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Rythmik L12 Subwoofer Review

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,835
Likes
16,497
Location
Monument, CO
I played around with accelerometers (Velodyne used that approach) and various other sensing methods before deciding upon a dual voice coil approach similar to what Rythmik does. I was in college at the time and it is not a very complex circuit; understanding the acoustic side was harder (and the subject of a grad class I took later). I still have my project in a nice little woodgrain (vinyl wrapped) metal box. It comprised an adjustable low-pass filter, high-pass filter (crossover), delay (all-pass) circuit, servo (LDI feedback) circuit, and a buffer to adjust the gain and bridge the amplifier driving the cone (several amps, but usually a Hafler DH-200 I had, and the speaker driver was a dual voice-coil Infinity IRS woofer and later a Radio Shack bass amp driver when a moving company destroyed my Infinity driver :( ).

Regarding "fast", as others said above to me it relates more to the ability of the sub to "start" and more importantly "stop" quickly. Back then a number of LF driver/amp combinations when "ring" for a while after the signal stopped, messing up the tail end (decay) of the transient. Most of the "fast" sound comes from frequencies well above the sub's range so what you "heard" was sort of a muffled woofing/pumping sound. There are various ways to solve that problem, as well as the problems of resonances and such, with servo being one way. Servo can also help compensate distortion and power compression (to a degree) and a little (analog or digital) EQ along with some safe operating area (SOA) circuits can often let a smaller sub play lower and louder than a comparable non-servo design.

FWIWFM - Don
 
Last edited:

Ron Texas

Master Contributor
Joined
Jun 10, 2018
Messages
6,078
Likes
8,914
@mitchco do you know the latency (delay) of the L12's?
 

Mnyb

Major Contributor
Forum Donor
Joined
Aug 14, 2019
Messages
2,641
Likes
3,618
Location
Sweden, Västerås
Do not Rythmik provide kits anymore , you can buy a plate amp with their servo tech and a suitable woofer from them .
And DIY a cabinet solution.
 

Alice of Old Vincennes

Major Contributor
Joined
Apr 5, 2019
Messages
1,418
Likes
903
I have the lvr 12 sub rear-ended firing port purchased a few years. The replacement fires from the front. I mated with the magnepans. Could not use room correction with planers. I was able to integrate pretty well by ear with the rear controls. I used high base extension and 24 db.
I grew tired of magnepans and am replacing with Infinity. I purchased the Anthem 520 despite the review in this forum. I wanted (coveted) anthem room correction. I also ordered Infinity during sale but have not yet received do to shipping mix-up. Hope ARC can integrate the sub. My room necessitates corner placement with the monitor and mains. I know I will likely need a second sub. I plan to slowly replace 5.1 system with Revel.
 

Alice of Old Vincennes

Major Contributor
Joined
Apr 5, 2019
Messages
1,418
Likes
903
I am still using Panasonic Plasma. Purchased the last year before discontinued. Will keep until breaks down. Then I suppose OLED.
 

Alice of Old Vincennes

Major Contributor
Joined
Apr 5, 2019
Messages
1,418
Likes
903
I think properly designed ported subs can sound as "fast" and "musical" as sealed. I believe SVS and Rhythmic know this. Most cheap subs are ported and have painted ported subs with bad reputation with audiophiles. Also, young kid in cars want distortion. I remember my son coming how and the foundation shook. The sound was ugly and painful. You can make either type of sub sound bad.
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,835
Likes
16,497
Location
Monument, CO
Ported subs roll off faster than sealed below their port tuning frequency (so less "room gain"). Distortion is often higher at and below the port tune frequency, but above that frequency, ported designs typically offer greater output and/or lower distortion for the same output as sealed. Sealed subs can have an advantage in transient response, less ringing/more damping, but that too can be countered by good design. The bad reputation ported subs have is mostly (if not all) IME/IMO from bad designs and/or overdriven subs that "chuff" from the port being overdriven.

FWIWFM - Don
 

Trdat

Addicted to Fun and Learning
Forum Donor
Joined
Sep 6, 2019
Messages
967
Likes
396
Location
Yerevan "Sydney Born"
I think my brain just broke.

I was gonna say the same thing, I still can't figure out if all the pages are the one schematic or different ones. Wishful thinking makes me hope that someone perhaps can one day explain it to me. But I think in reality ill be in my 60's till I manage to achieve an understanding of that diagram. A goal that needs to be accomplished either way.
 

Trdat

Addicted to Fun and Learning
Forum Donor
Joined
Sep 6, 2019
Messages
967
Likes
396
Location
Yerevan "Sydney Born"
Let's look at a woofer.

Experiment:

A "tick" as a time alignment tool
35Hz sine test tone
Fade in over one cycle so it doesn't "click" too much at the beginning of the wave.
Output via JBL LSR 308
First Trace - test signal
Second trace - the in-room response (with some EQ)
Third trace - the in room response - defeated the EQ in case that was making trouble.You see the tick, and its echoes in the room
Then the recorded "bass" tone. I wonder if this bass is slow or fast QUOTE]


Ray I appreciate your effort and your response. Either you have refrained from explaining cause you presume we understand or there is a hint of sarcasm I haven't picked up or better still you don't want to lead us to your opinion rather let us decide for ourselves. But, it is really difficult to put two and two together if the goal of the post has no conclusion. I can read impulse responses somewhat, and understand all the basics but its hard to totally comprehend the true goal of your post without a basic conclusion so I can match up the stats and images with what your trying to say. In any case, I will keep reading it like a stubborn fool till it slowly comes to me and it will.
 

Francis Vaughan

Addicted to Fun and Learning
Forum Donor
Joined
Dec 6, 2018
Messages
933
Likes
4,697
Location
Adelaide Australia
I was gonna say the same thing, I still can't figure out if all the pages are the one schematic or different ones. Wishful thinking makes me hope that someone perhaps can one day explain it to me. But I think in reality ill be in my 60's till I manage to achieve an understanding of that diagram. A goal that needs to be accomplished either way.
It is a problem with patents. That one isn't as bad as many, but the prose is always brain meltingly awful. A patent has to explain that the patentee knows wher this stis in relation to the state of the art - the prior art. That can get tiresome. Then the whole idea is explained in excruciating detail. And then with every variant the author can think of to avoid someone else patenting them.
If you want the tl;dr - look at figure 21. That is the the entire schematic for a working sub.
 

Hidde

Member
Joined
Mar 20, 2020
Messages
15
Likes
19
Easy. Here is all you need to know to make your own Rythmik clone:
https://patentimages.storage.googleapis.com/e0/16/07/15a06e212b83e4/US6104817.pdf
Calling servo or motion feedback easy is a severe understatement of it. It's probably one of the hardest parts to do right in loudspeaker design. Especially if you want to design a consistent controller (a controller that you can manufacture 10349193847 times).

You also need to have great knowledge about the operation of a driver and sensing mechanisms.

The patent you mentioned only shows it with an extra winding on the voice coil. Sure this works, but you get significant coupling between the signal you send and the signal you want to measure. Your signal will be disturbed.
The patent also limits itself to feedback only (no feedforward), analog only, and no adaptive mechanisms. I didn't read all the way through to see if they also use transconductance amplifiers.

The publications by Rob (of Grimm Audio and also professor mechatronics at Delft university) show a modern approach to servo/motionfeedback subwoofer design. If you really want to design your own servo sub I highly recommend reading those.
Here's a link: https://www.rmsacoustics.nl/audiodesign.html
 

Francis Vaughan

Addicted to Fun and Learning
Forum Donor
Joined
Dec 6, 2018
Messages
933
Likes
4,697
Location
Adelaide Australia
I was understating a trifle. :)
The point about the linked patent is it is that patent that covers the Rythmik subwoofers - it is by the company founder. It describes the design of the subwoofer reviewed here. Want a Rythimik clone, here is your resource.
The Grimm Audio link above has some really nice stuff, and covers another part of the ground nicely. Well written and very understandable. Some is still pretty blue sky. But, yes, does touch on feedforward. IMHO feedforward is the right way of doing an active sub. I'm not convinced a Kalman Filter is the best approach, but it is none-the-less one worth trying. I'm more inclined to try modelling the actual changeable parameters (ie, voice coil temperature, gap modulation) as use these to drive a parametrised model rather than hope a Kalman Filter can divine what is going on. But the reality is that I have no idea if one would turn out better than the other in practice. I'm still working towards a prototype breadboard system to try some ideas out.
 

QMuse

Major Contributor
Joined
Feb 20, 2020
Messages
3,124
Likes
2,785
Am i correct to assume if LFE input is used and PEQ is switched off that inputs signal bypasses all crossover/phase/EQ processing, that signal path is purely analog going only through amp and that only volume control is active?
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,835
Likes
16,497
Location
Monument, CO
Am i correct to assume if LFE input is used and PEQ is switched off that inputs signal bypasses all crossover/phase/EQ processing, that signal path is purely analog going only through amp and that only volume control is active?

Yes, LFE bypasses all but gain control, but the path is always purely analog.
 

QMuse

Major Contributor
Joined
Feb 20, 2020
Messages
3,124
Likes
2,785
Yes, LFE bypasses all but gain control, but the path is always purely analog.

I see, thank you. So in that scenario I can expect a sub delay to be pretty much equal to stereo amp+passive mains?
 
Last edited:

mitchco

Addicted to Fun and Learning
Audio Company
Joined
May 24, 2016
Messages
640
Likes
2,397
@mitchco do you know the latency (delay) of the L12's?

Hi Ron, no I have not set up an experiment to specifically measure latency of the L12.

In the context of a biamp or triamp system using digital XO, Audiolense accurately measures the delay relative to the other drivers. My article on integrating the Rythmik L12's shows the relative delay to the other drivers. Also included is a step response chart that shows the relative delay of about 3ms behind the mains, even though the mains and subs are pretty much in the same horizontal plane as the mains. This delay was also about the same for the F18's that were in the same spot as the L12's from the article. Note I use the line inputs and not the LFE. On the subs, the phase delay is set to 0, crossover max at 120 Hz, LPF slope none, bass extension - low music.

As a side note, the XLS 1502 also has it's own delay that I measured to be about 60 samples at 48 kHz, which is about 1.25ms. That's with everything bypassed. Given sound travels roughly a foot per millisecond puts context around the delay.
 

Ron Texas

Master Contributor
Joined
Jun 10, 2018
Messages
6,078
Likes
8,914
@mitchco I had seen those numbers before but incorrectly thought they were for the 18" subs. I wonder how much additional delay using a high pass on the XLS 1502 creates, given that the DSP is always active. My L12's are a foot closer to the listening position than the mains. I suspect that the subs are between 0 ms and 1 ms delayed. I will have to study your article again.
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,835
Likes
16,497
Location
Monument, CO
I see, thank you. So in that scenario I can expect a sub delay to be pretty much equal to stereo amp+passive mains?

I'd assume so but never measured it, sorry.
 
Top Bottom