Yamaha HS50M for hifi
In Japan it’s a popular practice to use mini studio monitors as the primary speakers; mainly due to lack of space. The NS-10M has become a legend not only in recording studios, but also in many homes of audiophiles. It has become the ‘industry’s standard’ so to speak, with its signature white cone.
I heard from many audio ‘experts’ that the NS-10M is extremely flat, which I disagree. The high frequency is pretty harsh so one can spot a poor recording quickly. A typical ‘monitor’ design. The 2 kHz range is highly boosted and revealing. This deliberate ‘equalization’ will help find fault in the midrange like vocals. If it sounds bad on the NS-10M, it mostly probably would be a very poorly mixed or equalized recording.
The mid rage is definitely very clear and analytical, and can be pretty uncanny for broadcasted voices like the BBC. In the old days of NS-10M, a matching amplifier was also a difficult choice. The Japanese old school audiophiles employed the JVC Victor AX5 with the old Denon DP40F. They consider that one of the classic combination in the 80s.
Now is 2011 and NS-10M is history. Here cometh the HS-50M which is an attempt to take the place of the NS-10M. Judging from the sales in US and worldwide these years, looks like it is indeed widely accepted! I started having a vested interest in the HS-50M after hearing them in Guitar Center, Natick MA. I finally bought a pair, after spending too much time fumbling about building a few clone LS3/5a.
I bought the HS-50M not for mixing, but to a have taste of what accuracy is about. I wanted a neutral speaker with no coloration. It could act as tool to spot poor recording from my vast collection of CD and DVD audio. Most of all, I like the sexy look of the white cones.
After using it for a while with computer equipped with a high end Creative sound card, I was somewhat disappointed about the sound of the HS50M. I am not a recording engineer and what the heck I bought it for? It sounded too harsh and flat, too clinical with no soul. Whether it was CD, DVD or DVD audio from the web, it sounded too thin and muddled.
I thought it was due to the limitations of the sound card, and used a good CD player and a simple pre-amp combo to drive the unbalanced input of the speakers. Still, no joy. Rated at 45W each speaker, the speakers can sound loud. But very painful.
As the HS50M cost much less than a pair of bookshelf + integrated amp, I was not too bothered and had the pair set aside for over a year. I had the intention to use it with my Clavinova organ. After listening to a pair of vintage NS-10M driven by NAD amplification, I thought the HS50M was pretty close to that NS-10M. Since I did not fully investigated the equalization setting of the HS50M, had I done it a great injustice?
Honestly, the HS50M sounds rough, unrefined for me. I opened the speaker back panel and considered some changes could further improve its smoothness and flavor as a hifi speaker.
The HS50M has an EI type transformer which is very close to the control electronics. I thought the shielding can be very poor and just next to the electronics circuit. This certainly could affect the sound quality. The input coupling caps are El cheapo polarized electrolytic bypass with a polyester small cap in the signal path.
Hence, after some consideration, I bought the following parts for upgrade:
A high quality 50VA 18-0-18 toroidal transformer to replace the original EI. I could reconfigure to 230V or 110V input as needed, using the new transformer. Should I need to bring the speaker into countries with 220-230V, I could still use this US model.
Nichicon MUSE non polar type hifi caps for the signal path. Original was small tiny 10uf 16v. I bought 6 MUSE 22uf 25v to improve signal integrity.
Original EI transformer, large, bulky and at 45VA (?).
New ILP or equivalent toroidal in comparison. Much higher efficiency, 50VA, and more compact. Much less vibration, and less humming.
The back plate showing the EI removed
Instead of drilling an ugly hole tru the back metal place, some high quality quick dry epoxy helps keep the mount screw secured.
When the glue for the mount shaft is setting, the coupling caps can be upgrade. The green MUSE caps can sound very accurate and musical too. Increasing to a larger value cap help adds bass definition.
Now we can change the signal path caps. As I only use the unbalanced option, changing the two caps should suffice. On the daughter board, the marked circle accepts the live AC power. Do solder the hot side of the primary winding to the location marked (1) as should in the right hand solder pads.
As the mount screw is set as a rock, the toroidal TX is mounted, with some additional glue as to secure everything.
There is no need to run the wire carrying acall over, under and close to the control board as the original wiring. See the markings. For 18V ac, 0 gnd and 18V ac.
Cable ties used to prevent vibration and rattling.
Check all connections, this is a simple mod. Close the back panel and screw back the screws. Do not forget to patch up the holes for the EL transformer mounting. If you are worried you made a mistake, try with a 2A fast blow fuse for testing.
It’s done! Tested with a good CD source carefully, the speakers just sound fine. The speaker can be driven by a compact DAC with headphone amp, such as the Go Vibe Magnum. But this Magnum’s flaw can be brutally revealed by the HS-50M, be warned. After hours of breaking in, the sound became smooth and pleasant and yet with the macro definition.
Now we can configure the speaker with the boost/defect controls. The MID EQ set to +2 would make it sound very much like a properly setup NS-10M, with little difference that is hard to tell. To make it sound more hifi, just set the MID to -2. Depending on your taste ( or the preferences of your ears), the high trim can be set to flat or +2.
The room control I set it to flat. For bass cut off, I set to 80 Hz or 100 Hz, don’t ask for too much form a 5inch cone.
So how does it sound?
The sound is much more more open and lively! Less muddled than what it was before using the same source and some configuration.
Setting it up for trying:
Overall, the HS-50m with the appropriate EQ setting, can sound very open and neutral, and more musical after the modification. The sound is clean but with more feeling. I suppose the toroidal with much lesser interference, vibration, and higher efficiency, made the little monitor perform much better, and hence more pleasant! I compared with a pair of stock HS-50m, the improvement was very distinct.
Can it play vocal? Absolutely. Can it play light jazz? You bet! These pairs of speakers are real bargains. The new toroid and coupling should do the magic. And they FAR cheaper to get a pair of LS3/5a, but I must admit the 3/5a has its own musical charm. Well, you at least do not need the amplifier for sure.
In Japan it’s a popular practice to use mini studio monitors as the primary speakers; mainly due to lack of space. The NS-10M has become a legend not only in recording studios, but also in many homes of audiophiles. It has become the ‘industry’s standard’ so to speak, with its signature white cone.
I heard from many audio ‘experts’ that the NS-10M is extremely flat, which I disagree. The high frequency is pretty harsh so one can spot a poor recording quickly. A typical ‘monitor’ design. The 2 kHz range is highly boosted and revealing. This deliberate ‘equalization’ will help find fault in the midrange like vocals. If it sounds bad on the NS-10M, it mostly probably would be a very poorly mixed or equalized recording.
The mid rage is definitely very clear and analytical, and can be pretty uncanny for broadcasted voices like the BBC. In the old days of NS-10M, a matching amplifier was also a difficult choice. The Japanese old school audiophiles employed the JVC Victor AX5 with the old Denon DP40F. They consider that one of the classic combination in the 80s.
Now is 2011 and NS-10M is history. Here cometh the HS-50M which is an attempt to take the place of the NS-10M. Judging from the sales in US and worldwide these years, looks like it is indeed widely accepted! I started having a vested interest in the HS-50M after hearing them in Guitar Center, Natick MA. I finally bought a pair, after spending too much time fumbling about building a few clone LS3/5a.
I bought the HS-50M not for mixing, but to a have taste of what accuracy is about. I wanted a neutral speaker with no coloration. It could act as tool to spot poor recording from my vast collection of CD and DVD audio. Most of all, I like the sexy look of the white cones.
After using it for a while with computer equipped with a high end Creative sound card, I was somewhat disappointed about the sound of the HS50M. I am not a recording engineer and what the heck I bought it for? It sounded too harsh and flat, too clinical with no soul. Whether it was CD, DVD or DVD audio from the web, it sounded too thin and muddled.
I thought it was due to the limitations of the sound card, and used a good CD player and a simple pre-amp combo to drive the unbalanced input of the speakers. Still, no joy. Rated at 45W each speaker, the speakers can sound loud. But very painful.
As the HS50M cost much less than a pair of bookshelf + integrated amp, I was not too bothered and had the pair set aside for over a year. I had the intention to use it with my Clavinova organ. After listening to a pair of vintage NS-10M driven by NAD amplification, I thought the HS50M was pretty close to that NS-10M. Since I did not fully investigated the equalization setting of the HS50M, had I done it a great injustice?
Honestly, the HS50M sounds rough, unrefined for me. I opened the speaker back panel and considered some changes could further improve its smoothness and flavor as a hifi speaker.
The HS50M has an EI type transformer which is very close to the control electronics. I thought the shielding can be very poor and just next to the electronics circuit. This certainly could affect the sound quality. The input coupling caps are El cheapo polarized electrolytic bypass with a polyester small cap in the signal path.
Hence, after some consideration, I bought the following parts for upgrade:
A high quality 50VA 18-0-18 toroidal transformer to replace the original EI. I could reconfigure to 230V or 110V input as needed, using the new transformer. Should I need to bring the speaker into countries with 220-230V, I could still use this US model.
Nichicon MUSE non polar type hifi caps for the signal path. Original was small tiny 10uf 16v. I bought 6 MUSE 22uf 25v to improve signal integrity.
Original EI transformer, large, bulky and at 45VA (?).
New ILP or equivalent toroidal in comparison. Much higher efficiency, 50VA, and more compact. Much less vibration, and less humming.
The back plate showing the EI removed
Instead of drilling an ugly hole tru the back metal place, some high quality quick dry epoxy helps keep the mount screw secured.
When the glue for the mount shaft is setting, the coupling caps can be upgrade. The green MUSE caps can sound very accurate and musical too. Increasing to a larger value cap help adds bass definition.
Now we can change the signal path caps. As I only use the unbalanced option, changing the two caps should suffice. On the daughter board, the marked circle accepts the live AC power. Do solder the hot side of the primary winding to the location marked (1) as should in the right hand solder pads.
As the mount screw is set as a rock, the toroidal TX is mounted, with some additional glue as to secure everything.
There is no need to run the wire carrying acall over, under and close to the control board as the original wiring. See the markings. For 18V ac, 0 gnd and 18V ac.
Cable ties used to prevent vibration and rattling.
Check all connections, this is a simple mod. Close the back panel and screw back the screws. Do not forget to patch up the holes for the EL transformer mounting. If you are worried you made a mistake, try with a 2A fast blow fuse for testing.
It’s done! Tested with a good CD source carefully, the speakers just sound fine. The speaker can be driven by a compact DAC with headphone amp, such as the Go Vibe Magnum. But this Magnum’s flaw can be brutally revealed by the HS-50M, be warned. After hours of breaking in, the sound became smooth and pleasant and yet with the macro definition.
Now we can configure the speaker with the boost/defect controls. The MID EQ set to +2 would make it sound very much like a properly setup NS-10M, with little difference that is hard to tell. To make it sound more hifi, just set the MID to -2. Depending on your taste ( or the preferences of your ears), the high trim can be set to flat or +2.
The room control I set it to flat. For bass cut off, I set to 80 Hz or 100 Hz, don’t ask for too much form a 5inch cone.
So how does it sound?
The sound is much more more open and lively! Less muddled than what it was before using the same source and some configuration.
Setting it up for trying:
Overall, the HS-50m with the appropriate EQ setting, can sound very open and neutral, and more musical after the modification. The sound is clean but with more feeling. I suppose the toroidal with much lesser interference, vibration, and higher efficiency, made the little monitor perform much better, and hence more pleasant! I compared with a pair of stock HS-50m, the improvement was very distinct.
Can it play vocal? Absolutely. Can it play light jazz? You bet! These pairs of speakers are real bargains. The new toroid and coupling should do the magic. And they FAR cheaper to get a pair of LS3/5a, but I must admit the 3/5a has its own musical charm. Well, you at least do not need the amplifier for sure.
Last edited:
hi Nimo, tks for your article!
I just bought a pair of 50M on Saturday. The sound is very sharp. Comparing to my Yamaha MSP3,which I have modified repeatedly for he last 3 years , MSP3 is far better. I have replaced the input op amp, add power supply capacitors, and chnge the electrolytics caps to Nichicon Muse 2, or by-passed them as they are too many in the circuitry.
in The 50M, I have by-passed the input caps after the balanced input, as well as the one to the volume control. from the volume pot, there are 2 plastic caps, I think they are for low feq cut. I took out one so far. I add MUSE 2 caps to the Op amp power supply and the power IC PS. too bad I dont hve the circuit diagram to the input/control board.
Wonder any body have the circuit diagram of the 50M circuitry.
I just bought a pair of 50M on Saturday. The sound is very sharp. Comparing to my Yamaha MSP3,which I have modified repeatedly for he last 3 years , MSP3 is far better. I have replaced the input op amp, add power supply capacitors, and chnge the electrolytics caps to Nichicon Muse 2, or by-passed them as they are too many in the circuitry.
in The 50M, I have by-passed the input caps after the balanced input, as well as the one to the volume control. from the volume pot, there are 2 plastic caps, I think they are for low feq cut. I took out one so far. I add MUSE 2 caps to the Op amp power supply and the power IC PS. too bad I dont hve the circuit diagram to the input/control board.
Wonder any body have the circuit diagram of the 50M circuitry.
Hi there, the circuit diagram will be the band pass, high/low pass filter configurations. The components are surface mounted so it may not be a worthwhile effort to replace them. The power amp section is biased into a class AB operation and is adequate.
The main reason for replacing the transformer was dual voltage operation (110V or 230V ac). The model I purchased only handles 110V. Unfortunately, the windings on the original transformer do not allow changing of voltages. This upgrade to toroids significantly improved the sound.
The rest of mods may not be apparent as the source could be a computer audio card or a professional mixer. If we replace everything then it will be a never ending story. As said, the monitor is used to expose the fault of the recording, and it is not easy to use the monitor for hifi.
Let the speakers break in for a couple of weeks or months. You will find the sound improve over time. The Japanese audio designers like to parallel a 0.1uf myler cap with a polarized 10~47uf capacitor for signal coupling. This produce a type of sound they like, and with a very low cost. A non polar cap can sound much better but with cost penalty. So replacing the input coupling caps already help improving the 50M in terms of signal integrity.
Also, I saw a recording guy using a pair of Furutech power cables and Hisago 6N PCOCC signal cables on the 50M. The cables together cost more than the speakers! And the speakers do sound great on some good recordings.
Last edited:
This pair of speakers have a rather uneven response curve with strong emphasis in the mid range.
Hi,
Could you just further explain what this means ? Which caps are they ?
The 22uF are now soldered, I'm juste waiting for the next step ...
Regards
Do you still know which input opamps you used for the MSP3? Do you have schematics?
Thanks
I tried to replace surface mounted op-amps before ( AD811 or AD812) . Although the wide band width and slew rates are very desirable, soldering SMT op amps needs SKILLS, tools, magnifying glasses and ... many more.
The PCBA track is weak, so this step is a big risk. replace the original NJM op amps is a good idea but a big risk.
The PCBA track is weak, so this step is a big risk. replace the original NJM op amps is a good idea but a big risk.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.