• 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!

Kenwood L-08M Monoblock Amplifier (Vintage Review)

GXAlan

Major Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
3,866
Likes
5,953
Kenwood L-08M Monoblock Amplifier Review
(Vintage Gear; QuirkAudio restored)
1663998949569.png


Where were you in 1980? In Tokyo, Japan, a small team of engineers released the most sophisticated amplifier of its time...

A bit of history
In the 1970's, an elite team of veteran Trio-Kenwood engineers left the company to form what is today known Accuphase. Half-rivals and half-colleagues, Kenwood would actually maintain a sizeable ownership of Accuphase all the way through the mid-1990's. Accuphase had the goal of building flagship audio equipment using the very best in craftsmanship and engineering. With a focus on quality not quantity, Accuphase continues today as a leader in high-end Class A and Class AB amplifiers and has a deep inventory and vault of spare components and parts for their electronics going all the way back to its founding.

Meanwhile, the young and talented engineers that remained at Kenwood wouldn't let their mentors take all of the glory to Accuphase-Kensonic Laboratory. At the start of what would we today recognize as Japan's "bubble economy" of the 1980's, an elite team at Kenwood set out to build the very best amplifier its engineers could achieve. With Accuphase-Kensonic already giving the Trio-Kenwood executives access to traditional high-end audio, the engineering team at Kenwood were not encumbered by the requirements of tradition or heritage. They were truly free to try out different concepts and ideas.

The L-08M represents the pinnacle of Kenwood's efforts in the audiophile arena never again to be matched on the specification chart. Let's look at the advertised specifications:

1663977065251.png


Want a high damping factor?

"How about 15,000!" Benchmark Media has a great blog explaining why that number should seem unbelievable -- speaker cables set a practical limit on damping factor. Kenwood engineers knew that too. To solve this issue, they had a second pair of wires going from the amplifier to the speaker terminals so that the negative feedback could be measured in such a way that included the effects of the speaker cable. This technology was called Sigma Drive.

Amplifiers need to run cool?
"We're going to make the whole chassis a sand casted heatsink." See that 27.5 lb weight? All heatsink.

Need a wide bandwidth?
"DC to 600 kHz!"

Want a fast slew rate to avoid TIM?
"How about 200V/microsecond!"

Want low distortion?
"We're going to apply negative feedback all the way to 50 kHz!"

Worried about crossover distortion?
"We'll avoid this with an ultra-high speed design!"

Worried about electromagnetic issues?
"We'll use as much non-magnetic material as possible! State-of-the-art material plastics!"

Repairability?
Why are you asking so many questions? Just look at how beautiful this is!

1664000312956.png


Maintaining a stock of old parts?
Our colleagues at Accuphase are stockpiling transistors for no good reason. We'll just order replacement parts from our suppliers whenever we need them, have them delivered to us via Pan-Am Cargo. With the way the semiconductor world is growing, I'm sure in 2 or 3 years, we'll be able to order anything we need to repair these from a Sears Mail Order Catalog.

(That's what they call foreshadowing.)

Very few specimens of the L-08 line still exist. Even rarer is my120/240V version with a German Type F plug. The plastic portions for the power switch and rectangular component of the amp and top-heavy beefy heat sink design makes it easily to topple or get damaged in transit. If the Sigma Drive cables are incorrectly wired, the whole unit is destroyed. If the remote trigger is wired to a 12V trigger, it probably destroys the unit too. This remote trigger was designed specifically for the L-08C and just wants the open circuit to be closed like a simple switch, as the entire trigger mechanism is self-powered by the amplifier itself. I'm sure user error adapting a 12V trigger to this also destroyed a lot of these amplifiers.

This particular unit was damaged during transit as it was shipped from Europe to the United States at the very beginning of the pandemic. It looked like the box was dropped, and it was clearly a stressful time for all of the world. It was repaired and recapped by QuirkAudio and in full disclosure, Peter from QuirkAudio has made additional recommendations to further improve its performance which haven't been applied yet. There's a real chance we can improve the performance even more.

Right now, Sigma Drive is operational on the L-08M, but I was cautioned by Peter that if the unit fails from a bad transistor while Sigma Drive is enabled, his opinion is that there is a real chance it will destroy components that just aren't available anymore and I'll end up with a beautiful heatsink. If I leave Sigma Drive off, it probably won't make a big difference in sound and the chances of catastrophic failure decrease substantially. Even the switch is apparently particularly complex. So... this unit is being tested with Sigma Drive disabled. We probably get even better performance with the feature enabled. And again, Peter from QuirkAudio still wants to make some tweaks some of the capacitors, but he and I have both been too busy. These amps see regular use driving my primary speakers in my home theater.

Oh, the marketing photos don't give you a true sense of the scale for the monoblock amplifiers. A pair of L-08M's takes more volume than the Marantz PM-10 and this photo comparing it to the svelte JBL SA600 gives you a good sense of scale.

1663990413284.png



Test Conditions
The signal chain is Panasonic UB9000 -- balanced-2V --> Marantz PM-11s2 as pre-amp --single ended--> Kenwood L-08m --Vishay Dale 1% NH-250 4.48 ohm resistor -- AUX in -> E1DA Cosmos
I have tested these in the stereo mode of the E1DA and ultimately used the AUX input for my final tests.

Initial testing idiosyncrasies...

I started with XLR input for 5W. I've blocked out the numbers so you can pay closer attention to curve and the ultrasonics in particular...
1663988430808.png


and then switched to the AUX to test higher power output
1663988457411.png

1663988485464.png


Whoa! Did you see that? What's going on at ~30 kHz and ~70 kHz? I went back to the original 5W tested power and back to using the XLR inputs again. No change -- I still had those spikes.
I took a break for dinner and then went back to testing again. This is where it gets really interesting. I just stuck with the AUX input on the E1DA Cosmos ADC.

1663988549394.png


1663988556425.png


1663988570999.png


See those humps at ~12 and 20kHz?

1663988578187.png


Did you see that? The humps at 12 kHz and 20 kHz dropped! The graph explains what I did. I turned off the lights in the room. LED lights really can affect signal quality. I did not have a dimmer on the lights so they were running full brightness as well. With the LED lighting out of the way, I continued increasing power.

1663988600773.png


1663988607956.png



It's pretty interesting to know that LED lights can introduce issues in the audible spectrum. Although the actual contribution is low, I can certainly imagine this playing a role. Now that I know that LED lights create this noise pattern, it gets even more interesting that different interconnect cables are more or less sensitive to the interference of the LEDs.

Oh, by the way, did you notice those numbers?

93.5 dB ~5W SINAD achieved 42 years ago in 1980 with high gain (1V input sensitivity). From Kenwood.
(And that's on my not-as-good-as-Amir's test setup. My PM-90 is 8-9 dB worse than his PM-90 amplifier measurements.)

As with the case with solid state, as you increase power, the noise goes down, allowing us to nearly reach 100 dB SINAD score at higher power. I am pretty sure I can improve the numbers if I fed the unit a better DAC with digital volume control. Right now, it's UB9000 then a PM-11s2 as pre-amp and then the amplifiers.


Subjective Impressions
Truth. Transparency. Makes bad recordings worse. Makes great recordings incredible. If I only had one amplifier, it'd be the Marantz PM-10 just for its versatility. If I am allowed to keep two amplifiers, my #2 would easily be the Kenwood L-08C/L-08M combo.

When I first brought the L-08M's home and fired it up, I was actually very disappointed. Vocals sounded thin and I just didn't enjoy it as much as my tube amp or warmer sounding Marantz PM-90 which measures worse. I was listening to the soundtrack from La La Land where you have actors/actresses singing (not necessarily professional singers). As I switched back and forth between the Kenwood L-08M and the Marantz PM-90 fed directly by a Topping D50s, I quickly was able to identify the difference. On the Kenwood, voices from the La La Land Soundtrack sounded like they were coming from something a little over an inch wide: a human mouth. With tube amps or the PM-90, the voices were larger than life as if I was watching something at a movie theater rather than having someone actually in the room.

Switching back and forth, it was pretty clear to me that the Kenwood was a more accurate amplifier that could bring Emma Stone into the room. But no matter how much my brain told me that I should want and prefer the absolute truth and transparency in my gear, I still personal have the preference for a larger-than-life sound and euphonic colorations that I get with the tubes or my Marantz gear -- at least for that specific soundtrack.

But classical music? Wow.
LP vinyl? Wow x 2.

Classical music is brilliant on this and it makes grand pianos sound authentic. I cannot describe exactly what is special and a null test might be a great way to assess this. I'm very familiar with piano music, and back when I was actively taking lessons, I could walk through a hotel or mall, hear an actual piano in the background and tell you with complete accuracy whether it was a Steinway, Kawai, Yamaha, or other. What really stands out with the L-08M is what I call micro-imaging for lack of a better term. Even with LP vinyls, you not only get a clear phantom center, a clear left and right image, but what separates this L-08M from basically every other amplifier I've listed to, other than the PM-10, is an uncanny ability for this amplifier to place instruments in sharp focus in a way unlike any other amplifier in the space IN BETWEEN phantom center and left or right. I have had other monoblocks such as those from Adcom and PS Audio 200c, but none of them achieved what the Kenwood could. I actually think the Kenwood L-08M beats the Marantz PM-10 for delivering the most narrow phantom center image (specifically for the La La Land Soundtrack) but the PM-10 gives me the chameleon of clean vocals with the tiniest amount of tube warmth with all of the crystalline clarity of classical music where direct comparisons can be made to reality.

The L-08M does great in my home theater setup. Harmonious Live on Disney+ sounds truly breathtaking with the L-08M. Even though the show is streamed in lossy Dolby Digital Plus, it indicates to me that the L-08M doesn't need some sort of fancy audiophile master recording. The L-08M is just a window of truth. It cannot make very good singers (who are better at acting) like Emma Stone and Ryan Gossling sound like spectacular singers the way that my Marantz gear can. The L-08M instead makes me feel like Simon Cowell -- suddenly I can easily hear the difference between world-class singers and very good singers.

Conclusion
More than 40 years ago, state of the art amplifiers like the Kenwood L-08M were driven by engineering and science. The designers behind this amplifier at Kenwood never gained the status of icons like Nelson Pass, Sidney Corderman, Frank McIntosh, Gordon Gow, Ken Ishiwata, John Siau or even the Accuphase founders, Nakaichi and Jiro Kasuga. As far as I know, they remain anonymous. But it is clear with my measurements that they deserve recognition in the history of hi-fi audio. They were not afraid to try new concepts or do things. Service manuals are readily available, and though many of the parts are now unobtainium, any patents are long expired.

This amp easily beats the average amplifier tested here and likely would have been the best measured amplifier of its time... two years before the introduction of the Compact Disc (in the USA). With Sigma Drive active, the negative feedback in real world loads would have been even better and this unit would have been fully transparent to even the 16-bit/50 kHz Soundstream recording technology of the era.

In testing this, I did discover the nugget of truth behind a lot of audiophile myths. Clearly something introduced those high frequency spikes during my testing, and this could be in the area where "AC line conditioners" could have made a difference. None of my other amps tested so far showed this kind of sensitivity to household electricity, so it's possible that older amplifier designs were more sensitive to "dirty power" at least in the realm of measurements.

I've attached PDFs of the technical explanation behind Sigma Drive and the original Kenwood L-08C/L-08M brochure. I don't remember where I sourced these, so apologize for not giving credit where credit is due.

Oh, and one more thing...

I tested the L-08M this evening yet again... when most of the neighborhood is asleep and most of the lights were off in my home.

1664002345283.png


1664002357246.png
 

Attachments

  • Kenwood L-08 brochure.pdf
    1.3 MB · Views: 441
  • Kenwood - Sigma-Drive Technical Guide.pdf
    4.1 MB · Views: 364
Last edited:
OP
G

GXAlan

Major Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
3,866
Likes
5,953
If I am interpreting this correctly, Kenwood, 42 years ago, was giving us pretty much HypeX NC500 performance.
Green dashed lines at 0 dB and -140 dB to show equal scale. orange line at -110 dB.

The Kenwood is giving us this performance at 1V input sensitivity on an unbalanced connection too with a UB9000/PM11s2 in the signal chain!

1664005827343.png
 
D

Deleted member 48726

Guest
Surely the noise from the supply can't travel through the power supply and the filters of the amplifier, -right? It must be from (interfering through) the source?
 
OP
G

GXAlan

Major Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
3,866
Likes
5,953
Surely the noise from the supply can't travel through the power supply and the filters of the amplifier, -right? It must be from (interfering through) the source?

No idea. I just know that turning on and off the LED lights reproducibly creates and removes those humps at ~12 khz and 20 khz. The 30 kHz and 60 kHz spikes popped up during testing and are gone in the dead of night.

The ~12-20 kHz humps are HIGHER with higher output power which suggests a source issue. I had a previous post that shows that interconnect cables made a difference.

The 30 kHz and 60 kHz spikes appear to be the same no matter what the output power is. That's what would be interesting to figure out.

Again, the chain is
USB stick --> Panasonic DP-UB9000 -- Canare L-4E6S / Neutrik Silver balanced XLR --> Marantz PM-11s2 -- StraightWire or Monster Cable 800 THX unbalanced interconnects --> Kenwood L-08M

The Kenwood amps have grounding lugs which are used to ground the E1DA Cosmos ADC.
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,579
Likes
38,274
Location
Gold Coast, Queensland, Australia
The LO series were quite an achievement. I had 9 (yes nine) LO-7 monoblock power amps (both mk1 and mk2s) along with LO7C and a LO7Cmk2.

The LO7m/mk2 rated at 150wpc, same other specs delivered 190wpc@8R and 289wpc@4R on my bench. I sold my last pair about 10 years ago as I simply wasn't using them. (sometimes regret it, but you can't keep everything.)

There is a separate transformer for the trigger supply and it is pretty much bulletproof and the Sigma Drive takes a lot more than incorrect wiring to cause trouble. Disconnection at high power however is not pretty.

Your L08M likely has the Toshiba MT200 30MHz/15A transistors, not the LAPT/RET style used in the L07ms.

Does your L08m have arguably the best RCA implementation ever with screw collar RCAs on the special OFC leads or did you not get those? Designed to prevent someone tripping over the long leads and making a hell of a noise in the amps. I got these direct from Japan back in the day for the Kenwoods and never made up shorter pre-power amp leads. (the originals were 10M or more AND directional...).

screw on rca med.jpg
 
Last edited:

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
Whoa! Did you see that? What's going on at ~30 kHz and ~70 kHz?
Please make this test: send the signal to amp input, but the measuring cable at amp output would be shorted and connected to the amp Lo (gnd) output binding post (black). You should measure the straight line at the position of system noise. Anything else would be a mesurement error.
Re noise floor in multitone measurements, this may be a question of the window used.
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,579
Likes
38,274
Location
Gold Coast, Queensland, Australia
Please make this test: send the signal to amp input, but the measuring cable at amp output would be shorted and connected to the amp Lo (gnd) output binding post (black). You should measure the straight line at the position of system noise. Anything else would be a mesurement error.
Re noise floor in multitone measurements, this may be a question of the window used.

I agree, it's definitely a measurement error. I had the same thing drive me crazy at my bench a few years back, albeit at about 16kHz and multiples. Never got to the bottom of it. Wasn't USB packet noise, wireless, LEDs, SMPSs, plugpacks, phones, even mouse and keyboard wasn't the problem. No idea what caused it but it wasn't the HiFi as it appeared across all sorts of gear. Sampling rate made no difference except 'sometimes' moved it up, not always...
 
  • Like
Reactions: pma

charleski

Major Contributor
Joined
Dec 15, 2019
Messages
1,098
Likes
2,239
Location
Manchester UK
It's pretty interesting to know that LED lights can introduce issues in the audible spectrum. Although the actual contribution is low, I can certainly imagine this playing a role.
Um, those humps are down at around -130dB. Although it's an interesting demonstration, I can't imagine them having any audible effect. I think it's safe to listen to music with the lights on ...
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,579
Likes
38,274
Location
Gold Coast, Queensland, Australia
Um, those humps are down at around -130dB. Although it's an interesting demonstration, I can't imagine them having any audible effect. I think it's safe to listen to music with the lights on ...

It becomes an exercise in "where the f#$k is that coming from??". Trust me, strange spectral spikes drive us up the wall... ;)
 
D

Deleted member 48726

Guest
No idea. I just know that turning on and off the LED lights reproducibly creates and removes those humps at ~12 khz and 20 khz. The 30 kHz and 60 kHz spikes popped up during testing and are gone in the dead of night.

The ~12-20 kHz humps are HIGHER with higher output power which suggests a source issue. I had a previous post that shows that interconnect cables made a difference.

The 30 kHz and 60 kHz spikes appear to be the same no matter what the output power is. That's what would be interesting to figure out.

Again, the chain is
USB stick --> Panasonic DP-UB9000 -- Canare L-4E6S / Neutrik Silver balanced XLR --> Marantz PM-11s2 -- StraightWire or Monster Cable 800 THX unbalanced interconnects --> Kenwood L-08M

The Kenwood amps have grounding lugs which are used to ground the E1DA Cosmos ADC.
Well, LED drivers PWM frequencies are all over the place, some in the hundreds, others above 20.000 Hz so it could be fun disassembling one of your lights and look up the specific switching frequency of the driver. -But again that may be to go to the extremes. I know I have enough "brain malfunction" to not letting it go if it was me..:facepalm:o_O
 
OP
G

GXAlan

Major Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
3,866
Likes
5,953
Does your L08m have arguably the best RCA implementation ever with screw collar RCAs on the special OFC leads or did you not get those?
Sadly it does not. Did not have the remote cables either. I did get an aftermarket 2 meter Sigma Drive cable to connect the L-08C to L-08M.


Um, those humps are down at around -130dB. Although it's an interesting demonstration, I can't imagine them having any audible effect. I think it's safe to listen to music with the lights on ...

Oh, what I meant was the <20 kHz frequency. The ceiling lights are pretty far away and my home was built (electrically) less than 10 years ago, so it’s got great cables. I’d have to wonder if a dimmer, bad light, older wiring, etc. might get you worse problems. The other issue is that we say that bad anti aliasing filters can trigger intermodulation distortion that is both audible in amplitude and frequency

But yes, the effect as tested here would easily be masked by the real signal.
 
OP
G

GXAlan

Major Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
3,866
Likes
5,953
I agree, it's definitely a measurement error. I had the same thing drive me crazy at my bench a few years back, albeit at about 16kHz and multiples. Never got to the bottom of it. Wasn't USB packet noise, wireless, LEDs, SMPSs, plugpacks, phones, even mouse and keyboard wasn't the problem. No idea what caused it but it wasn't the HiFi as it appeared across all sorts of gear. Sampling rate made no difference except 'sometimes' moved it up, not always...

What’s weird is that I didn’t touch anything between the three different measurements. In the first round where the high frequency noise was introduced, all I was doing was turning the knob on the preamp to increase volume.

Then later, when everyone was asleep, I just retested without changing anything and the noise was gone!

Just wasn’t sure if ~30 kHz was a classic tell-tale ________.

Since the THD+N measurements are just 20-20kHz, it doesn’t affect the numbers much other than a fraction of a dB lower noise

It’s impressive that Kenwood achieved this performance which is transparent to CD quality five years before the advent of CDs.

I am not sure what transistors are in the L-08M. The service manual lists the transistor type but not the manufacturer.

91188562-B212-4C6A-AE15-8CEAA31C3766.jpeg
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,579
Likes
38,274
Location
Gold Coast, Queensland, Australia
What’s weird is that I didn’t touch anything between the three different measurements. In the first round where the high frequency noise was introduced, all I was doing was turning the knob on the preamp to increase volume.

Then later, when everyone was asleep, I just retested without changing anything and the noise was gone!

Just wasn’t sure if ~30 kHz was a classic tell-tale ________.

Since the THD+N measurements are just 20-20kHz, it doesn’t affect the numbers much other than a fraction of a dB lower noise

It’s impressive that Kenwood achieved this performance which is transparent to CD quality five years before the advent of CDs.

I am not sure what transistors are in the L-08M. The service manual lists the transistor type but not the manufacturer.

View attachment 233063

Toshiba:

(Internet pic of the L08M OPT set)
1664065235130.jpeg
 

FrantzM

Major Contributor
Forum Donor
Joined
Mar 12, 2016
Messages
4,335
Likes
7,719
One further proof and reminder of how great the Japanese designs of the 70 and 80's were. Across the board. It pains to realize how great a disservice the High End Adio press TAS, S'Phile and their ilks , did to the hobby and industry in general.
I would like similar tests/reviews for some , generally unknown (Japanese) speakers of the same period. We know a lot about Yamaha efforts (NS-1000, etc) but little else about others, in particular Technics... They made several SB-xxx "Linear Phase", some of which, were said to be superb.
Congrats and thanks to the OP @GXAlan

Peace.
 
Last edited:
OP
G

GXAlan

Major Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
3,866
Likes
5,953

Doodski

Grand Contributor
Forum Donor
Joined
Dec 9, 2019
Messages
20,745
Likes
20,756
Location
Canada
Great review @GXAlan and nice to see a Kenwood tested here @ ASR. I've handled some Kenwood high end gear and they where truly built like a tank and had premium pots and stuff. This one looks really nice. Can we see some pics of the interior and the rear plate or what would usually be the rear connection plate.
 

Chrise36

Major Contributor
Joined
Dec 23, 2018
Messages
1,043
Likes
359
Surely the noise from the supply can't travel through the power supply and the filters of the amplifier, -right? It must be from (interfering through) the source?
It does i have tested my old Technics amp with a speaker driver and i hear more noise when devices with smps are working. The more smps the more the noise.
 
Top Bottom