Sehlin Helium DIY Speaker Review

[Vini darko]

That impedance graph is looking pretty bad. So bad its shown twice

 

[Mudjock]

I'll address a few of the questions and comments as I not only submitted the Helium for review, but also am the designer.

Is the graph a measurement or a simulation?

The answer is "a little bit of both". The driver frequency responses were directly measured in the box. The crossover is simulated. What was posted came out of my working design files.

Is that a 4 inch woofer? Or smaller? You didn't give those details. Also a view of the rear would be nice. I suppose we also don't have any similarly small speakers to compare it to do we.

Any ideas on how that jagged response from 800 hz to 2 khz happens?

The woofer is the Dayton ND91-4, which is actually around 3". The jagged response from 800 Hz to 2 kHz is something we haven't totally resolved. I have the other speaker of the pair and took some measurements on that speaker after Amir shared the spin.

This measurement was taken at a similar SPL (about 80 dB at 1 meter). The microphone was at tweeter height 0.5 m in front of the baffle. This has no post processing other that a pretty shot gate time of about 3.5 msec. In designing this, I couldn't really solve the dip around 1 kHz with a passive crossover that would fit in a box this small, but was able to reasonably balance everything else out on axis. We don't see the jaggedness from 800 Hz to 2 kHz and we see the tweeter level in better alignment with the "woofer" level. I debated not publishing this review, but ultimately decided that transparency was the better course of action as this forum has science in the title and maybe some insight could be gained as there were some aspects of the measurements that did align. I couldn't completely rule out that those things that didn't align could be design related.

If I had to guess at the source of the jaggedness in Amir's measurements, the two most likely causes seem to be either some odd sounds coming from the port (which is facing the front) that my measurement process is gating out, or damage to the woofer (either from shipping or testing). I normally test at a lower level than what Amir tested at. When I brought the level up to 80 dB at 1 meter, the woofer was trying pretty hard to jump out of the box at the low end of the chirps (20 Hz). I didn't even try 86 dB, which is Amir's standard level and the first thing he at. These can play that loud with music signals, but 20 Hz has the 3" woofer unloading pretty severely.

That impedance graph is looking pretty bad. So bad its shown twice

The original impedance curve shown on the website is smoothed and based on a crossover simulation. The actual woofer impedance I took at that time didn't have good resolution at low frequencies (It was one of the first measurements I took with Praxis and I didn't have all of the settings sorted out). I took one recently to compare with Amir's and it matched quite closely. This was an important data point to establish that the crossover in the speaker Amir measured was intact.

Totally not what we have here. We have narrowing directivity centered at 5kHz and a widening of directivity centered at 9kHz. Dayton shows wide directivity at 5kHz and a slight narrowing at 7kHz.
I wonder if this is just how unit-to-unit variance is for a <$10 tweeter, or if there is something else at play.

I was going through my files and found an off-axis family of measurements I did about 6 years ago (0, 15, 30, 45, and 60 degrees off axis). It seems to confirm Amir's measurement in that region. I am not sure if that measurement came from the speaker I sent in, or the one that stayed here, so I can't comment on unit-to-unit consistency without further investigation.

Sehlin Helium is a fair attempt to create cheap, very small loudspeaker with easy to find, off the shelf ingredients. How to compare? Opponent should be small also, with comparable price. Not easy to find! Jeff Bagby Quarks?

It's interesting you mention these. We were doing a small speaker comparison at a DIY event in Indiana a few years ago. Jeff Bagby and I were sitting next to each other. The Quarks and Heliums played back-to-back and we both preferred the Heliums (at least in that venue).

My near-field workstation area is in pieces so I initially tested the Helium far field. Well, let me just say the sound is not right. Tonality is wrong (muffled and boomy), and there is grittiness to sound which I did like. I then sat in front of it while still on my normal stand and the situation improved. But the grittiness was still there.

This is a little telling. This design has been around quite a bit (several large DIY gatherings and probably well over 100 builds) and I have never heard them described as "gritty". I would normally associate gritty sound with a defective or damaged driver. This is something I will investigate further when I have the tested speaker back in hand.

I thank Amir for the measurements and review. These were never the flattest measuring speakers, but the design has been generally well received. There were some unexpected anomalies in the measurements here. The end result is that it may raise more questions and answers, but hopefully there is something to be learned through the exercise.

 

[PeteL]

So very very small, it's interesting but they are not really portable, I think that the use cases are a bit limited for something like this no? I'm just wondering why would be SO limited in space to have to go for something as tiny on his desk, especially unpowered. Personally I own a pair of Yamaha MSP3 that are not that much bigger than this, if bigger.. A bit more expensive, but not really since they are amped and "labour" free. To me they sound OK, and would likely measure better than this but even still I don't use them to really listen to music daily, they are a tool that I use rack mounted for monitoring applications in professional applications, and they are great for that, but it's still something limited, as a permanent nearfield system I would, I do, use something else (my everyday computer speakers are ADAM F5, I do have a quite better main system). What I'd like to see more here, if we talk about very small speakers, would be to see measurments and reviews of the small portable Bluetooth speakers like JBL Charge and Flip serie, or the competition from Bose and UE. Yes, you can laugh, but those have a way more defined use case, I have a Charge, and I DON'T expect Hifi, but it doesn't matter, I'm able to shut the audiophile analytical part of my brain and have that to enjoy a glass of wine and good music outdoor, anywhere, with some friends, if we want to see what very small can do, those are way more useful in everyday life. I'd be curious to see how bad it really is.

 

[wwenze]

Desk space can be a luxury in some rooms. If I get bigger than 27" screen I will have to downsize my current speakers.

We need some sort of innovation in this area. Under-monitor, behind monitor, speaker array of 1-inch drivers etc.

 

[hardisj]

Get the $90 Neumi BS5 that @hardisj measured. Thanks to him it’s now going to my de facto budget recommendation over the previous king of the budget market; the Micca MB42X.

Glad to hear that. I'd be curious what your thoughts are. Feel free to PM or share in the other thread. Getting additional subjective opinions is always helpful and I'm curious if you had similar impressions to mine or if we varied, and how we use/d them.

 

[ROOSKIE]

Love the DIY. The simple fact is that you can build a pair of Parts Express C-notes for around the exact same price (USA prices at least).
They really are not very large and put out substantial volume and even decent bass.

If you really need tiny speakers for a small space/desk/wall mounting ect, IMHO a better idea would be to go with small sealed enclosures and then partner with an appropriate sub. Use active high pass as well.

 

[ROOSKIE]

Howdy, since this is a small speaker thread, as a side note the KEF EGG active sets are on sale for $199 at KEF direct in the USA. If you need a small set-up such as the DIY speaker featured here and like KEF you may want to buy a set for cheap...

https://www.shop.us.kef.com/egg-wir...ISVAHOL6V1Bg3p5VNzo1QYcOV-BIw908aAviPEALw_wcB

 

[JeffS7444]

What I'd like to see more here, if we talk about very small speakers, would be to see measurments and reviews of the small portable Bluetooth speakers like JBL Charge and Flip serie, or the competition from Bose and UE. Yes, you can laugh, but those have a way more defined use case, I have a Charge, and I DON'T expect Hifi, but it doesn't matter, I'm able to shut the audiophile analytical part of my brain and have that to enjoy a glass of wine and good music outdoor, anywhere, with some friends, if we want to see what very small can do, those are way more useful in everyday life. I'd be curious to see how bad it really is.

Unfortunately it never occurred to me to keep the data from my bluetooth speaker measurements, but I recall seeing pretty good on-axis frequency from JBL Charge 3 at close range and moderate volumes, ditto a very tiny cube-shaped Sony, but in the case of the Sony, you practically had to hold it up to your ear to appreciate it! Most challenging were the old Bose Soundlink and Soundlink Mini 2 where high frequencies were pretty rolled off by 10 kHz, and I noted a +5 dB boost at 50 Hz - perhaps the Bose speakers would have fared better outdoors.

 

[BYRTT]

How did you place the microphone for 0°? Was it tweeter axis offset to the center of the tweeter, or center of the cabinet?..... .....Totally not what we have here. We have narrowing directivity centered at 5kHz and a widening of directivity centered at 9kHz. Dayton shows wide directivity at 5kHz and a slight narrowing at 7kHz.
I wonder if this is just how unit-to-unit variance is for a <$10 tweeter, or if there is something else at play.

.....I was going through my files and found an off-axis family of measurements I did about 6 years ago (0, 15, 30, 45, and 60 degrees off axis). It seems to confirm Amir's measurement in that region. I am not sure if that measurement came from the speaker I sent in, or the one that stayed here, so I can't comment on unit-to-unit consistency without further investigation.....

MZMK in CAD sofware i bench XO region to be high up at 5,4kHz area as seen below animation tilt of spindata in verticals to point down at floor and up at cieling and think that explain case that woofer is dictating directivity curve all the way up to there, and imagine all the raginess or resonance in directivity curve for woofer is caused by a very near port leaking and baffle loss plus diffraction interference that will change with distance and angle, maybe Mudjock can confirm if XO region is that high..

EDIT many thanks to Amir for review..

 

[digicidal]

MZMK in CAD sofware i bench XO region to be high up at 5,4kHz area as seen below animation tilt of spindata in verticals to point down at floor and up at cieling and think that explain case that woofer is dictating directivity curve all the way up to there, and imagine all the raginess or resonance in directivity curve for woofer is caused by a very near port leaking and baffle loss plus diffraction interference that will change with distance and angle, maybe Mudjock can confirm if XO region is that high.

I wouldn't find that surprising at all considering we're talking about a "woofer" that's smaller than many midrange drivers employed in 3-way designs.

 

[JohnBooty]

On the subject of small speakers...

I'm curious if anybody here has heard the sound on the 2019 16" MacBook Pros.

The sound coming out of those things is remarkably clear. They have a two-way speaker system with six drivers, including two sets of dual-opposed (!) woofers.

A friend and I tested his 16" MBP with some frequency sweeps and there was (barely) audible output down to 60hz although, obviously, things rolled off rather steeply below 100hz. I'm sure there was active EQ at work, and obviously the max SPL was not large.

Still, I felt it was a pretty remarkable achievement for tiny speakers.

I've heard a number of small and large Bluetooth speakers and I've always found them barely tolerable at best, even when using the 1/8" input jack in order to avoid the woes of Bluetooth audio codecs. The 16" MBP was different to me. It was listenable.

 

[JeffS7444]

Wondering how Helium would fare for cinema use, especially for L, R, height and surround, where room correction is applied by the receiver? And how they'd compare to other very small speakers such as Cambridge Minx, Klipsch Chorus, B&W M-1 and so on.

 

[JohnBooty]

Wondering how Helium would fare for cinema use, especially for L, R, height and surround, where room correction is applied by the receiver? And how they'd compare to other very small speakers such as Cambridge Minx, Klipsch Chorus, B&W M-1 and so on.

I've never achieved satisfactory subwoofer integration with tiny speakers that have such high bass rolloff frequencies.

Keep in mind that with a typical 12dB/octave crossover slope, a subwoofer crossover of 100hz means your subwoofer will be reproducing audible output up at 200hz. That is localizable and right smack dab in the middle of the human vocal range. Steeper crossover slopes cure some issues and cause others.

Many feel otherwise, so I suppose I'm doing it wrong or imagining things.

 

[RayDunzl]

Desk space can be a luxury in some rooms. If I get bigger than 27" screen I will have to downsize my current speakers.

We need some sort of innovation in this area. Under-monitor, behind monitor, speaker array of 1-inch drivers etc.

I have two 30" monitors on my late 70's vintage rescued-from-the-recycler-switch-to-fancy-cubicles-at-work steel "tanker"desk.

My desk sound is a pair of Sony "Active Speaker System SRS-150"

I bought them in 1992 or so to go along with a Discman.

They are behind the left monitor.

I've never examined them too closely, nor measured them. They are unique (to me) in that the driver appears to be a square flat piece of white styrofoam with a rubber surround.

I don't think they are a two-way. The brief manual says "full range driver", 60~15kHz and 10% THD.

And just look at those snazzy dual ventilator chrome plated megaphone of destiny reflex loading ports! I know you're all jealous now.

They make sounds not at all unlike music when required.

*plays the clip above

The main system is in the same room if I need a little plus de qualité.

 

[digicidal]

I've never achieved satisfactory subwoofer integration with tiny speakers that have such high bass rolloff frequencies.

Keep in mind that with a typical 12dB/octave crossover slope, a subwoofer crossover of 100hz means your subwoofer will be reproducing audible output up at 200hz. That is localizable and right smack dab in the middle of the human vocal range. Steeper crossover slopes cure some issues and cause others.

Many feel otherwise, so I suppose I'm doing it wrong or imagining things.

That's always been my experience with satellite+sub setups. If there's no authoritative bass to at least 80-100Hz then they all sound like a BOSE lifestyle system... regardless of who is actually manufacturing them IMO. Although I wouldn't call 200Hz "the middle of the human vocal range" - maybe you hang out with 350lb lumberjacks regularly so our experiences may vary. I guess fundamental for speech is around there though so fair enough... you just won't be hearing that over any phone or most other recorded voices.

 

[ZgSTar]

I had a pair of Bowers & Wilkins MM-1 that didn’t sound too bad, and easily fit in the palm of my hand. I would think it possible to design something that tests reasonably well.

MM1 are still the best sounding mini desktop speakers I've heard. They are stably on my desk since a few years now. What a soundstage.

 

[mrmoizy]

How did you place the microphone for 0°? Was it tweeter axis offset to the center of the tweeter, or center of the cabinet?
______
Here are Dayton’s published measurements for the tweeter:
View attachment 79513
Totally not what we have here. We have narrowing directivity centered at 5kHz and a widening of directivity centered at 9kHz. Dayton shows wide directivity at 5kHz and a slight narrowing at 7kHz.
I wonder if this is just how unit-to-unit variance is for a <$10 tweeter, or if there is something else at play.

As far as I understand, infinite baffle measurements for individual drivers certainly have their purposes, but the outcome of taking that tweeter, putting a crossover in front of it, and putting it on such a small box will drastically change the measured outcome, including directivity.

 

[JohnBooty]

maybe you hang out with 350lb lumberjacks regularly so our experiences may vary.

Haha. If only my life was that exciting. Yeah, I was thinking about the fundamental frequencies of speech, which are so important in home theater applications.

 

[Tom C]

MM1 are still the best sounding mini desktop speakers I've heard. They are stably on my desk since a few years now. What a soundstage.

Mine blew a few years ago.

 

[PeteL]

Unfortunately it never occurred to me to keep the data from my bluetooth speaker measurements, but I recall seeing pretty good on-axis frequency from JBL Charge 3 at close range and moderate volumes, ditto a very tiny cube-shaped Sony, but in the case of the Sony, you practically had to hold it up to your ear to appreciate it! Most challenging were the old Bose Soundlink and Soundlink Mini 2 where high frequencies were pretty rolled off by 10 kHz, and I noted a +5 dB boost at 50 Hz - perhaps the Bose speakers would have fared better outdoors.

Thanks, mine says Charge 2+, it's ok for what it is I believe