• 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!

Klark Teknik Air Link Wireless System Review and Measurements

GXAlan

Major Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
4,492
Likes
6,949
Klark Teknik Air Link DW20T/DW20R Quick Review and Measurements
516rIwEP71L._AC_SL1000_.jpg

This is a quick review of the Klark Teknik Airlink DW20T/DW20R. This was purchased by me for wireless subwoofer use although it is intended for full range use. The unique feature of this system is that it has XLR balanced in/outputs. The combo runs about $200.

51EOcem4TmL._AC_SL1000_.jpg
51hws0A+mSL._AC_SL1000_.jpg

The transmitter is rated at handling up to 10 dBU (2.45V rms) and the receiver is rated at producing up to 15 dBU (4.36V rms). However this is at an advertised performance of 0.1% THD+N (60 dB SINAD)!

Official Specifications
Frequency response 20 Hz to 20 kHz (±1 dB)
THD + N (distortion) 0.1% @ 1 kHz
Signal-to-noise ratio >90 dB
Transmission frequency 2.406 - 2.475 GHz
Wireless range Up to 30 m (100 ft)
Transmission power 12 dBm
16-bit/48 kHz bit-depth/sampling rate
~14 ms latency
You need to re-link the devices if you lose power.
One transmitter can send the same stereo signal to 4 receivers.

Reliability
I bought an open box, looks-new receiver with Sweetwater warranty and had massive drop-outs. After discussing with the retailer, they shipped me a brand-new receiver unit! The replacement unit worked perfectly. So, if you buy one used/demo, make sure there is a trial period to make sure it's not defective.

Test Setup
The wireless transmitters were about 15 feet apart from each other in a home environment with mesh Wi-Fi 6. What makes testing this difficult is that the two units have an analog attenuator knob. Having the input too high can cause distortion even though it's lower than the 10 dBU advertised spec and doesn't show the red "clipping" light. I quickly settled on putting the receiver at maximum volume and the transmitter at 50% volume. This roughly gave me unity gain such that 2V going into the transmitter resulted in 2V going out from the receiver. It is likely that the transmitter is attenuating and the receiver is amplifying but it wasn't easy to set the knobs evenly while walking back and forth 15 feet each time. I then tried attenuating the volume on both, with the knob at the 9 o'clock position which was successful in reducing the high frequency distortion. I ran the initial set of tests using my Monolith HTP-1 outputting at 2V.

Measurements
You can see in magenta that there was a lot of noise introduced at the higher frequencies when running at input settings that I chose, but attenuating both resulted in the orange line. There is likely some ideal setting between the transmitter and receiver if you are running this full-range and I was likely running into some sort of clipping or potentially wireless interference.
1701208885669.png


This translates into this distortion profile for the volume setting #1 vs #2
1701208980822.png

1701209006068.png



Here are the measurements at the louder volume setting.

1701209127324.png

1701209242574.png



Compared to the REL
After doing my measurements, I re-read @amirm 's review of the REL wireless subwoofer. I tried to match his 30 Hz at 1.6V input, targeting 1.8V output. Using the HTP-1 setting of 1.6V nominal, I adjusted the input gain until the signal was flickering between green/red (green = signal, red = clipping) and then adjusted the receiver until I got 1.8V out. This was about the 1 o'clock position on the transmitter.

SINAD is 87.3 which beats the 74.4 dB of the REL device. Running the input to a solid green instead of flickering red/green had a SINAD in high 86 dB's so this was a bit better.
1701218976014.png


It might roll off the bass slightly more than the REL, but -0.5 dB is ~10 Hz or so instead of 6 Hz. This may also be the E1DA Cosmos ADC or the HTP-1 as signal generator.
1701219433161.png

1701219471558.png


Distortion is impressively low as a subwoofer wireless transmitter.

I then tried running a 2.4GHz wifi speed test while running Multitone 30 Hz at 128 repeats and didn't see a big difference. My laptop was in between the transmitter and receiver but lower, but the access point was in the same room about 10 feet away and out of line of sight.
1701219599884.png

1701219719118.png




Conclusion
The official spec is 0.10% THD+N, and in my testing, a 1 kHz test tone was 0.01% THD+N. When used as a wireless music transmitter to active PA speakers which often have high-gain, you're typically using voltages that are pretty low and do not run into clipping. Certainly, your mileage may vary when it comes to distance. Figuring out the right input/output levels is a challenge since the analog potentiometer is very sensitive.

Still, for my intended use as a wireless subwoofer transmitter/receiver, this is great. The bass frequencies are where the unit seems to work the best. Looking at the 60 Hz test tone, we can see that the first harmonic at 120 Hz is -106 dB which is pretty good and for the 30 Hz test tone, the harmonics are also in the range of -95 dB or better. Most subwoofers have much more distortion down low.

While this unit seems better than the REL, it's hard to do an apples-to-apples comparison since the test environment is different. Ergonomically, losing the pairing of the two when you lose power is annoying and while the antennas improve reception, the units are less attractive than the sleek black unit of the REL. That said, as far as I know, this is the cheapest XLR enabled wireless transmitter that seems suitable for wireless subwoofer duties.

EDIT: I ran this in stereo mode. I will test the transmitter in mono mode to see if that improves anything in the future.
 
Last edited:
Interesting review.
Thanks for that.

To set the volume, I'd try using an IMD SMPTE test signal.
I'd set level to +4dBu (balanced) or -10dBV (unbalanced), typical level out of the generator (DAC).
I'd set the receiver volume at a level that is low enough to avoid saturation at any input level, then play with the transmitter level until best IMD+N result. If no saturation is shown even with attenuator at 0dB on the transmitter, I'd then push the generator level just below saturation.
Then tune the receiver volume for the same: max level before saturation or best IMD result.

I'd use an IMD test signal because they have a low frequency main component and typically saturate lower than a single tone.
 
Last edited:
Very interesting, specially your use case transmitting sub only signal.
Knowing nothing about how these devices work, will the signal that reaches the sub stay always synchronised with the mains? As it seems it is intended for mono transmission too, I guess it should, just wondering.
And btw, what is the point of the mono knob? Does it sum the channels?
Thanks.
 
Very interesting, specially your use case transmitting sub only signal.
Knowing nothing about how these devices work, will the signal that reaches the sub stay always synchronised with the mains? As it seems it is intended for mono transmission too, I guess it should, just wondering.
And btw, what is the point of the mono knob? Does it sum the channels?
Thanks.

It should. I am reinstalling my home theater this weekend (hopefully) and will run the measurements. As long as the latency is relatively consistent then Dirac room EQ will handle everything correctly.

The mono switch lets you feed one signal and transmit the same thing to two signals. I wonder if it improves signal resiliency somehow if it’s sending a single channel as opposed to a single channel and self noise.

The marketing pitch is
“the convenience of high-performance audio-streaming to any installed or portable sound system, making it ideal for hospitality and event spaces, as well as retail, home audio, and much more.”

It seems like the main use are things like outdoor events.

Wireless subs have the convenience of letting you place subwoofers in portions of the room that would otherwise be hard or unsightly to run wires from the AVR/AVP to the subwoofer. The majority of subwoofers just have unbalanced outputs so you can use any number of wireless subwoofer kits. Mine happen to use XLR inputs and it’s nice to minimize ground loops.
 
  • Like
Reactions: MCH
What about the most important bit: the wireless range? I have a similar product from a “reputable” brand (can’t remember), but it barely makes it across the room without connection issues.
 
What about the most important bit: the wireless range? I have a similar product from a “reputable” brand (can’t remember), but it barely makes it across the room without connection issues.
They advertise 100 ft line of sight and run on 2.4 GHz which should have better resiliency than 5+ GHz bands. That said, there's a risk of WiFi interference although I did not see it in my particular environment.

I'm using it about 15 feet which works well. Not sure how best to test this, other than the fact that this is one of the few wireless transmitters that has an external antenna (Dynasty Pro Audio does as well). You pretty much need to buy this from a retailer that allows in-home trial since the RF environment is so different for everyone.

Did you happen to test latency? I tested three other wireless transmitters and found a huge range in latency. I see the spec is 14.

I took my slowest sub (Velodyne DD10) which is 0 ms for Dirac (likely due to the DSP). I moved that sub behind my sofa taking it from a 13 ft to 2 ft distance (give or take). That gives me 10 ms of latency reduction just by moving my subwoofer closer. Then. I re-ran Dirac and see about 3-4 ms more delay. My microphone isn't perfectly in the same spot between measurements, so the ~14 ms seems to be legitimate +/- 1 ms or so.

1701241147054.png
 
That said, as far as I know, this is the cheapest XLR enabled wireless transmitter that seems suitable for wireless subwoofer duties.

Thomann are selling the Sirus Stereo Link 5.8 Bundle for £111, they want £172 (£95 + £77) for the DW20T+DW20R....


I'm not sure how comparable it is, here's the specs from the manual...

1701270424439.png


 
Great device performance wise, but the gain/trim pot would disqualify it for me for any kind of professional use. I'd much rather see a 3-way sensitivity switch. Maybe one with 1:1 passthrough, one with 10dB gain and one with a 10dB pad or something like that would make it much more fool- and accident-proof.
Thanks for the review!
 
I'm not sure how comparable it is, here's the specs from the manual...

View attachment 330320

It looks so similar physically and on the spec sheet that I would imagine it’s the same ODM or OEM. More latency but 24-bit instead of 16 bit. The recommended and max voltages are interesting since that may explain why


Great device performance wise, but the gain/trim pot would disqualify it for me for any kind of professional use. I'd much rather see a 3-way sensitivity switch. Maybe one with 1:1 passthrough, one with 10dB gain and one with a 10dB pad or something like that would make it much more fool- and accident-proof.
Thanks for the review!
Agree. Or even something with a click center detent.
 
Thanks for sharing your test results with the community!
 
Measurements
You can see in magenta that there was a lot of noise introduced at the higher frequencies when running at input settings that I chose, but attenuating both resulted in the orange line. There is likely some ideal setting between the transmitter and receiver if you are running this full-range and I was likely running into some sort of clipping or potentially wireless interference.
View attachment 330140

If this was measured using log-swept sine or similar (Dirac impulse etc) then this could be a measurement error, not noise from the unit. Typically caused by the unexpected latency and the phase difficulties that brings to the analyzing software. There might be setting options to let the analyzer read the signal correctly (in AP these settings exist and often have to be adjusted to avoid these exact same artefacts, even in low latency situations).
 
If this was measured using log-swept sine or similar (Dirac impulse etc) then this could be a measurement error, not noise from the unit. Typically caused by the unexpected latency and the phase difficulties that brings to the analyzing software. There might be setting options to let the analyzer read the signal correctly (in AP these settings exist and often have to be adjusted to avoid these exact same artefacts, even in low latency situations).
Those measurements were just done in REW, so it very well may be a measurement error. I probably can re-run this using the frequency sweep mode of Multitone to see what happens with the higher signals.
 
S
Did you happen to test latency? I tested three other wireless transmitters and found a huge range in latency. I see the spec is 14.
They reason why I don’t use wireless headphones while playing Call of Duty. Wireless is too slow.
 
I don’t use wireless headphones while playing Call of Duty. Wireless is too slow.

+1. I use a digital piano and the latency from a setup that converts everything to digital before going back to analog can be enough to interfere with the responsiveness. It’s OK with active speakers with a 2ms of latency though.

14 ms added latency is really good for bass though. The fancy Rel’s are 17 ms.

What is interesting is that Sirus brand that was posted a few comments earlier is 24/48 and quotes 17 ms latency while this one quotes 14 ms and is 16/48. I would imagine that the lower bit depth helps to speed the audio data frames.

I originally was moving one of my subwoofers behind my sofa for aesthetics but I am surprised by the enhanced experience. The shorter distance and extra tactile response from the vibrating back of the sofa. I run DLBC and have full-range bed layer so the localization of feeling the vibrations is offset by audible bass that is not localizable and the bass remains accurate from a phase/level standpoint despite the increase in body feel. People talk about building nearfield subwoofers on AVSForum and I get it.
 
16-bit/48 kHz bit-depth/sampling rate

Most of the streamer now 24 bit/96khz. Probably need to be above that, 24bit/192khz is enough, to compete.
 
16-bit/48 kHz bit-depth/sampling rate

Most of the streamer now 24 bit/96khz. Probably need to be above that, 24bit/192khz is enough, to compete.

The trade off is latency. No reason to wirelessly transmit what isn’t audible. I imagine it would even be better if this was focused on subwoofer use because you could filter out the high frequency to reduce noise. You might even be able to decrease the sampling rate as well.

But I think it’s easier to sell a full range product.
 
The trade off is latency. No reason to wirelessly transmit what isn’t audible. I imagine it would even be better if this was focused on subwoofer use because you could filter out the high frequency to reduce noise. You might even be able to decrease the sampling rate as well.

But I think it’s easier to sell a full range product.
Yes, I found the multiroom such as Airplay 2 is useful. But it only max 48khz sampling rate.
Right now I prefer enjoy music come from several speakers wirelessly which is Apple HomePod to several Airplay2 speakers (using Airport Express mini Toslink), this can be soundbars or full system (DAC, pre-amp, amp, and speakers).
 
Last edited:
I have this pair and I just wanted to add that, since they use the 2.4 GHz band, the transmitter causes interference which makes some of my 2.4 GHz Wi-Fi devices behave strangely (connect/disconnect/slow down). I have not done measurements or collected other objective proof, however - as soon as I turn on the DW20T, these issues start, and they disappear immediately when it is powered down. Just FYI.
 
Back
Top Bottom