Erin Reviews Chesky LC1

[Heinrich]

Boy, these are rough. ...
- Extremely bad pair matching

Exemplary bad in every aspect, but no way ever superseeds Wilson TuneTot. The Prince, still unrivaled contender for the throne.

 

[miguelito]

I think you will find the answers to several of your points at Erin’s website. Also, Erin performs his tests using a Kippel, which measures the speaker in 360 degrees using industry standard methods ( hence results are directly comparable to other tests) and reports same per these industry standards. This way we can do apples to apples comparisons with other speakers.

To be clear: I am not challenging the tests he performs, they are probably showing the correct information.

However, what I am challenging is the validity of those tests when you're assessing the frequency response of a speaker in a medium sized room. ANY speaker in a medium sized room, but particularly a speaker that is by design energizing the room with reflections through the side passive radiators. If we were talking about, say, very directional horn speaker design, then I can see the on-axis anechoic response will be representative (to some degree) of the real life sound, but for speakers that are designed to interact with the room heavily like these, that is no longer true.

Additionally, there are inconsistencies in the review. For example, he says at some point the speakers cannot be broken because they measure the same (a reasonable argument) but then goes on to say that they don't measure the same in the crossover region. So what is it? Do they measure the same or not?

The dip in the crossover region he measured indicates that THAT speaker has the tweeter leads reversed - it could have happen at the manufacturing stage, or more likely he inadvertently reversed them while reassembling the speaker. In any case, seeing this discrepancy between the two speakers should have immediately prompted him to check. Such issue would also explain his impression that the speaker cannot center focus.

 

[sigbergaudio]

To be clear: I am not challenging the tests he performs, they are probably showing the correct information.

However, what I am challenging is the validity of those tests when you're assessing the frequency response of a speaker in a medium sized room. ANY speaker in a medium sized room, but particularly a speaker that is by design energizing the room with reflections through the side passive radiators. If we were talking about, say, very directional horn speaker design, then I can see the on-axis anechoic response will be representative (to some degree) of the real life sound, but for speakers that are designed to interact with the room heavily like these, that is no longer true.

But the test is not only measuring on-axis, it is measuring 360 degrees around the speaker.

And I don't think it's the case here that the inconsistencies on-axis is there deliberately to help it work in a room (?).

 

[dedobot]

I disagree with this notion - they're typically way denser harmonically and have much more low end than typical "audiophile music".

The typical sparse jazz and documentary style classical recordings are much more forgiving IME.

If a an speakers can reproduce this CD well they would also cope with large philharmonic orchestra ...and vise versa :

https://soundcloud.com/theclassicmixcdseries%2Fmiles-hollway-elliot-eastwick-in-at-the-deep-end

 

[AnalogSteph]

But then the same person says:

Speakers probably won't be much better than the Kali LP-6v2 for that little money, even though there are a lot of alternatives on the market with similar performance. With woofers for about 5 USD per speaker, you have to put up with some distortion.

About 5 USD per driver? I didn't know they were quite that cheap. That makes GRS drivers seems downright fancy.

That 6PT-8 as a midrange with, as you say, 15" woofer or equivalent, let's say a powerful one, plus some compression driver, there is the potential for a fun high SPL party speaker. Even a speaker with "nice" Hifi sound (good FR not just high SPL) if it is designed/constructed correctly.

Others have considered the 6PT-8 as a midrange, in a three-way design:

I want a cost effective passive 3-way xover to mate GRS 12PT to GRS 6PT plus "whatever" tweeter

the system would be a small (L100 size) reflex with woofer to midrange having good poer handling and hopefully a clear midrange, I'm having a lot of pain issues but wrose, sision issues so cannot see my monitor well (looks washed out to the point that light graphs in hornresp disappear). FWI I...

www.diyaudio.com

Ha, the 12PT actually is a near-perfect match. (But so would the 18PT be. Or a Visaton W300.) And managing to marry a 6.5" and 1" tweeter sans much of any waveguide with next to no warts deserves some accolades.

The 8PT-8 appears to be the only stinker with major unevenness in this line, I wonder what happened there. (BTW, they all seem to have Kapton voice coil formers. In a driver as cheap as the 6PT-8, that's remarkable. This also explains why it has higher power handling than the 6PR.)

 

[DanielT]

About 5 USD per driver? I didn't know they were quite that cheap. That makes GRS drivers seems downright fancy.

It could be a typo. Even if Kali were to buy up large quantities in bulk to get a low unit price, so ... well, it still seems too low a price. Anyway, the person who wrote what I quoted is familiar with the speaker manufacturing industry.

Here's what he said in the same thread:

You can assume that the production cost is about 1/4 of the consumer price in the EU (about the same VAT now), i.e. about 50 USD per speaker. In addition to overhead (including development) and profit margin for the OEM/ODM supplier, this should be enough for the driver, electronics, box, connectors, assembly, packaging and shipping. So here, every cent counts and driver+electronics probably won't cost more than about 16-17 USD.

faktiskt.io • Visa tråd - Att tillverka högtalare själv eller köpa färdiga?

Ha, the 12PT actually is a near-perfect match. (But so would the 18PT be. Or a Visaton W300.) And managing to marry a 6.5" and 1" tweeter sans much of any waveguide with next to no warts deserves some accolades.

The 8PT-8 appears to be the only stinker with major unevenness in this line, I wonder what happened there. (BTW, they all seem to have Kapton voice coil formers. In a driver as cheap as the 6PT-8, that's remarkable. This also explains why it has higher power handling than the 6PR.)

It seems like the potential for fun DIY speakers with a reasonable price.

 

[a4eaudio]

If you want a good, inexpensive DIY speaker using the GRS PR-8, then build this...

Cabrini Redux

midwestaudio.club

Or better yet, build the 2025 version using the newer GRS subwoofer at a stupidly low price that includes CNC'ed baffles...

Capacitor

18-160vY

www.jfcomponents.com

 

[levimax]

With so many choices for much better performing speakers at this price point (and below) I am surprised anyone bothers to give these a second look. I am not sure why anyone cares that a 17 year old with a famous last name designed them or that they have $5.00 woofers. The end result is suboptimal at best.

 

[AnalogSteph]

If you want a good, inexpensive DIY speaker using the GRS PR-8, then build this...

Cabrini Redux

midwestaudio.club

Looks oldschool, but solid enough. Measurements confirm fairly solid distortion performance of this driver.

Or better yet, build the 2025 version using the newer GRS subwoofer at a stupidly low price that includes CNC'ed baffles...

Capacitor

18-160vY

www.jfcomponents.com

8" woofer plus 28mm dome with a 2 kHz XO... hmm. Should work decently but I still wonder how it looks off-axis, and why he dropped the off-center tweeter.

Basic article | Baffle geometry

About baffle design, edge diffraction, secondary sound sources and their influence on the reproduction quality of loudspeakers.

heissmann-acoustics.de

Impedance of the 8SW-4 has more than doubled by 2 kHz, I imagine electrical nonlinearity won't be too great at this point. I normally wouldn't want to run that driver much past 1 kHz. No wonder it's being advertised as a sub.

 

[Cbdb2]

Strategically very - interesting - to praise the greateness of a product which is objectively not great at all.
This shows how untrustworthy Cheskies marketing claims are. Sad, but - lesson learned.

We expect that from manufacturers, its the Stereophool review that is the sad part. I new these reviewers were full of BS and pretend to hear differences with cables but when there really is a large difference they still cant hear it.
Whats with the passive radiaters that do nothing?

 

[dfuller]

8" woofer plus 28mm dome with a 2 kHz XO... hmm. Should work decently but I still wonder how it looks off-axis, and why he dropped the off-center tweeter.

Offset tweeters are intended to reduce baffle diffraction effects by basically making the tweeter a different distance from every edge, thereby reducing severity. This is at least in theory not a bad design choice in that department but it does screw up your horizontal beam some vs a center mounted tweeter, and (more importantly in this case, I'd gather) it makes manufacturing more complicated - now you need two different baffles rather than one.

Whats with the passive radiaters that do nothing?

well, they don't do nothing - they actually cause the screwed up response. Check this out. This is a WinISD simulation that matches up almost perfectly to the measured response. I don't have the exact internal dimensions so this is a guess at ~10.5L.

Interestingly, if you get rid of one of the PRs, the response actually improves some. Still not good, per se, but you're starting to creep towards an EBS.

 

[kemmler3D]

Compares favorably against competitors costing twice as much.”

There are enough overpriced sonic catastrophes out there that you can pretty much always say this, even if you don't bother to check.

 

[a4eaudio]

... and why he dropped the off-center tweeter.

Offset tweeters are intended to reduce baffle diffraction effects by basically making the tweeter a different distance from every edge, thereby reducing severity. This is at least in theory not a bad design choice in that department but it does screw up your horizontal beam some vs a center mounted tweeter, and (more importantly in this case, I'd gather) it makes manufacturing more complicated - now you need two different baffles rather than one.

The first speaker I linked to is a VERY inexpensive DIY speaker so mfg costs didn't come into play. It was designed over a decade ago and had an offset tweeter. This reduces diffraction ON AXIS (as the Heissmann link shows) but can worsen things off axis and make stereo imaging worse. It is a still a good choice on wide-baffle speakers, like a classic monkey coffin.

The second link was designed this year - drivers, xo parts, and CNC'ed baffles for $104.

[Note, I am not the designer, I just knew the speaker in the first link used the GRS woofer used in the Chesky speaker and people mentioned DIY builds.]

 

[MAB]

Whats with the passive radiaters that do nothing?

So true! You get this predicted change in output between the two in this box:

I guess the improvement is after you EQ the peak away save amp power...

The drones drive the woofer cone to large excursion at 35 Hz even with just over 10 Watts (input is 10 Volts, driver is 8 Ohms, xmax is 6mm). Here is the excursion of the woofer cones in the passive radiator in the same small cabinet.

This is a sealed woofer being unfortunately shoehorned into a passive radiator application, in a small box, with the drones actually upsetting the resonant characteristics badly.

The distortion even tracks the simulation for cone excursion in the passive radiator alignment, which is not surprising if I got the simulation details right.

The woofer cone of this speaker is really going to be jumping around in real-world use, even with bass shy music.

 

[dfuller]

This is a sealed woofer being unfortunately shoehorned into a passive radiator application, in a small box, with the drones actually upsetting the resonant characteristics badly.

I'm like, 90% convinced it has PRs because the designer thought they looked cool. No shade, he's literally a kid. A little bit of simulating with free software (VituixCAD is free, WinISD is free, etc etc) would pay off in spades.

 

[kyuu]

I'm like, 90% convinced it has PRs because the designer thought they looked cool.

Probably also because "PRs are free extra bass, just like a port except better!"

 

[AnalogSteph]

I guess the improvement is after you EQ the peak away save amp power...

The concept strikes me as something better suited to an active speaker anyway. With a steep highpass below 40-50 Hz, excursion would actually remain quite reasonable.

I guess the PRs could use some more mass?

I'd also give the old series capacitor trick a shot, it does tend to help the peaking with drivers shoehorned into small enclosures. Maybe 470µ?

 

[JD_Spoon]

The dip in the crossover region he measured indicates that THAT speaker has the tweeter leads reversed - it could have happen at the manufacturing stage, or more likely he inadvertently reversed them while reassembling the speaker.

I have confirmation that the tweeter leads were not reversed in the units that were measured.

 

[MAB]

The concept strikes me as something better suited to an active speaker anyway. With a steep highpass below 40-50 Hz, excursion would actually remain quite reasonable.

I guess the PRs could use some more mass?

I'd also give the old series capacitor trick a shot, it does tend to help the peaking with drivers shoehorned into small enclosures. Maybe 470µ?

Yes, the capacitor trick would help this quite a bit.
Since 8 Ohm, slightly smaller value capacitor:

It does become useable, with DSP slightly more so.

This woofer more suited to sealed and is going to be on a knife-edge with passive radiators, especially in this size box.

 

[dfuller]

The concept strikes me as something better suited to an active speaker anyway. With a steep highpass below 40-50 Hz, excursion would actually remain quite reasonable.

I guess the PRs could use some more mass?

I'd also give the old series capacitor trick a shot, it does tend to help the peaking with drivers shoehorned into small enclosures. Maybe 470µ?

Nah, the driver itself is not suited to bass reflex enclosures. I've modeled it out, it never really works except in a sealed box. Even then it isn't anything amazing. Ultimately we're looking at a driver more suited to use as a midrange than as a woofer.