• 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!

Buckeye Hypex NC252MP Amplifier Teardown

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
45,610
Likes
252,362
Location
Seattle Area
I just reviewed the 6-channel Buckeye Hypex class D amplifier. This is a follow up teardown to that review.

The amplifier is an assembly affair as all active electronics are contained in the Hypex NC252MP stereo class D amplifiers with integrated switching power supplies:

Buckeye Amplifier 3XNC252MP teardown DIY.jpg


So the main job is routing the wiring to each module and input/output terminals. I was pleasantly surprised by the quality of the construction. While I will comment on some areas of improvement, they are really minor.

AC power input is from top right through the fuse and routed using the twisted read and white wires. While this coloring scheme is consistent with DC wiring, it should not be used for AC wires. Otherwise someone working on them may confuse them with DC low voltage wires with catastrophic consequences. The convention when AC and DC circuits are present in the same device as is the case here, is to use black or brown for "hot" and white or blue for neutral. The grounding terminal likewise uses a black wire and should instead use solid green or green and yellow wire:

Buckeye Amplifier 3XNC252MP Grounding.jpg


I was pleased to see that the anodization was sanded off where the screw terminal is for safety ground. An ordinary washer is used. For belts and suspenders purpose, it is better to use a barbed washer that has "teeth" that really dig into the metal.

Plastic double sided taped clamps hold the wires (e.g. mains). These come loose over time and should be avoided. There are screws near all the wires so I suggest using a screwed in clamp. That said, I can't see anything bad happening if they come loose as they are not going to touch any sharp/hot metal.

Throughout the unit I found fine aluminum shavings/bits. Many were tiny but some were large enough to dislodge and cause a short in the fine pitch surface mount PC boards. You can see the specs in my larger shot above on the right near the speaker/XLR terminals.

Speaking of those terminals, they are all screw mounted which rules out me critiquing the soldering job. :) All were tight and tidy:

Buckeye Amplifier 3XNC252MP teardown XLR speaker terminals.jpg


Using adhesive lined shrink tube is a very nice touch and above norm for such builds or even commercial ones. The adhesive keeps air out and should keep the tubing in place past my lifetime probably.

Zooming in a bit into Hypex modules we see the same practice of using second or third tier electrolytic capacitors:

Hypex NC252MP capacitor.jpg


There are large vents on top that should keep the ambient temps low inside the unit and help with longevity of the unit. Note however that the main source of heat are hidden power transistors that couple to the bottom of the chassis. Don't put the amplifier on top of any other heat generating component. I can't tell if there are heatsink compounds between the amps and the bottom plate. If not, some should be there.

Conclusions
I usually expect the worst when an audiophile decides to build such amplifiers before the teardown. I was super happy to see that some of the best safety and signal integrity rules have been followed here leaving me with some minor nits. While the overall unit has no safety or regulatory certification (although the modules are likely such rated), I personally would not be concerned about using this amplifier in my own system.

-----
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
 
Thanks! The minor suggestions are easy enough for me to incorporate (washers, colored wire, etc).

Other than I now blow out the cases before closing up to clean out even tiny shavings, the one issue I still run into is the front panel screws are thread forming (just the way the company I buy from assembles the cases) so when I close it up at the end, the screws sometimes still dislodged tiny shavings from the extruded aluminum sides during tightening.
I am working on seeing if they can switch over to machine screws or to find my own that work to avoid this.
 
Low-voltage wire (LED wiring) touching single-insulated AC wire. :)

Common practice is to use double-insulation all the way until the connection *if* there is chance of touching DC.
If physical isolation can be guaranteed (i.e. wiring secured properly with sufficient spacing between AC and low-voltage), single-insulation can be used.

This is good resource: https://www.cui.com/catalog/resource/power-supply-safety-standards-agencies-and-marks

For power supplies, the minimum insulation requirements are:
  • Primary to secondary: Reinforced insulation with a minimum dielectric strength of 3000 Vrms.
Protective Earth
Class I equipment must have a protective earth conductor, if insulated it must be a green/yellow or transparent covering. No switch or fuse is allowed. Resistance between earthed parts and the earth termination must not exceed 0.1O, which is tested by a current 1.5 times the current capacity of any hazardous voltage circuit at the point where failure of basic insulation would make the earthed part live. Test voltage maximum is 12 V. Test current may be ac or dc but must not exceed 25 A.
Clearances (through air)
Power products are intended for general application and should be designed for worst case conditions (pollution degree 3 and mains voltages up to 264 Vac) for minimum clearances of:
  • 4.00 mm for reinforced or double insulation.
  • 2.00 mm for basic and supplementary insulation.
Where formal quality control processes are in place relaxation to 3.4 mm and 1.7 mm is allowed, but reinforced insulation is then subjected to 100% electric strength testing. If an air gap serves as the insulator between a hazardous voltage and the enclosure, the required clearance is 10 mm.
 
Last edited:
Do the NC252s need (or benefit from) an input buffer?
 
A few more thoughts.
First up, wire colour - yes - please use the proper colours. (Being in Oz, we follow the EU colours, brown and blue for active and neutral, and stripped yellow/green for safety ground. I really don't like other standards, as black and white get used for other things.) No matter what you go with, don't use the colours for anything else.

You are using red and black inconsistently. You really need even more colours. The standby output supply is taken on a black wire out, and returns from the relay as red. Don't use black either. It gets generally recognised as negative or zero volts. Here it is +5v.

You have not used the standby power grounds. So the second and third amplifier boards are having to reference their PS Enable signal via the ground used for input shield. Don't do this. There is, of necessity, current flowing in the input shield if this happens, and it is as dirty as the standby power is, plus there is a large loop formed inside the cabinet which can inject noise. Connect the standby ground pins of all the amplifiers together. Yes all the grounds are connected together somewhere in the PCBs, but that doesn't absolve you from ensuring that the ground returns are correctly managed, and loop area minimised. There is a very specific set of rules followed in layout of the boards that manages the ground return currents, and forcing current through the wrong path you compromise the noise immunity of the design. This alone could be the cause of the disparity in mains noise Amir measured. Once amplifiers acheive the performance we get from Hypex, every detail counts, as it is trivially easy to just slip up on one and noticably compromise the performance.

As Amir says, don't use stick on cable tie mounts. They will inevitably fall off.

For general niceness, don't tie other cables in with a mains bundle. Get the front panel LED runs out of the mains run.
 
Do the NC252s need (or benefit from) an input buffer?
These modules already have buffers in them and don't need them. But that does not prevent companies like PS Audio from adding yet another buffer before their buffered modules (Stellar series use ICEpower) to add their "flavour".
 
Last edited:
Glad to know my buckeye amp (not this particular amp) passes the Amir tear down test. Good job buckeye.
 
Thanks Amir ad Buckeye, looks neat! What about those self adhesive wire ties? I see a number of people using them, and while they seem to only maintain a path for power, it would think that when they eventually come loose (heat from the boards), that path will no longer be maintained. In a solderless system, perhaps that's a concern?
 
You have not used the standby power grounds. So the second and third amplifier boards are having to reference their PS Enable signal via the ground used for input shield. Don't do this. There is, of necessity, current flowing in the input shield if this happens, and it is as dirty as the standby power is, plus there is a large loop formed inside the cabinet which can inject noise. Connect the standby ground pins of all the amplifiers together. Yes all the grounds are connected together somewhere in the PCBs, but that doesn't absolve you from ensuring that the ground returns are correctly managed, and loop area minimised. There is a very specific set of rules followed in layout of the boards that manages the ground return currents, and forcing current through the wrong path you compromise the noise immunity of the design. This alone could be the cause of the disparity in mains noise Amir measured. Once amplifiers acheive the performance we get from Hypex, every detail counts, as it is trivially easy to just slip up on one and noticably compromise the performance.

Thank you for all the tips. This one I am still a little confused on. Are you referring to Pin 4 on J6 (GND) for each module? The way I have them wired now is that Pin 3 (Vout Standby) from the first module is connected to the 12v relay then split coming off the relay going to each modules PS Enable. This way of wiring I confirmed with Hypex was OK (essentially one modules Vout Standby can be used to control the PS Enable of multiple modules).

Are you saying on top of this wiring scheme, I should also be connecting Pin 4 (GND) of J6 of each module together as well?
 
I just reviewed the 6-channel Buckeye Hypex class D amplifier. This is a follow up teardown to that review.

The amplifier is an assembly affair as all active electronics are contained in the Hypex NC252MP stereo class D amplifiers with integrated switching power supplies:

View attachment 100587

So the main job is routing the wiring to each module and input/output terminals. I was pleasantly surprised by the quality of the construction. While I will comment on some areas of improvement, they are really minor.

AC power input is from top right through the fuse and routed using the twisted read and white wires. While this coloring scheme is consistent with DC wiring, it should not be used for AC wires. Otherwise someone working on them may confuse them with DC low voltage wires with catastrophic consequences. The convention when AC and DC circuits are present in the same device as is the case here, is to use black or brown for "hot" and white or blue for neutral. The grounding terminal likewise uses a black wire and should instead use solid green or green and yellow wire:

View attachment 100588

I was pleased to see that the anodization was sanded off where the screw terminal is for safety ground. An ordinary washer is used. For belts and suspenders purpose, it is better to use a barbed washer that has "teeth" that really dig into the metal.

Plastic double sided taped clamps hold the wires (e.g. mains). These come loose over time and should be avoided. There are screws near all the wires so I suggest using a screwed in clamp. That said, I can't see anything bad happening if they come loose as they are not going to touch any sharp/hot metal.

Throughout the unit I found fine aluminum shavings/bits. Many were tiny but some were large enough to dislodge and cause a short in the fine pitch surface mount PC boards. You can see the specs in my larger shot above on the right near the speaker/XLR terminals.

Speaking of those terminals, they are all screw mounted which rules out me critiquing the soldering job. :) All were tight and tidy:

View attachment 100589

Using adhesive lined shrink tube is a very nice touch and above norm for such builds or even commercial ones. The adhesive keeps air out and should keep the tubing in place past my lifetime probably.

Zooming in a bit into Hypex modules we see the same practice of using second or third tier electrolytic capacitors:

View attachment 100590

There are large vents on top that should keep the ambient temps low inside the unit and help with longevity of the unit. Note however that the main source of heat are hidden power transistors that couple to the bottom of the chassis. Don't put the amplifier on top of any other heat generating component. I can't tell if there are heatsink compounds between the amps and the bottom plate. If not, some should be there.

Conclusions
I usually expect the worst when an audiophile decides to build such amplifiers before the teardown. I was super happy to see that some of the best safety and signal integrity rules have been followed here leaving me with some minor nits. While the overall unit has no safety or regulatory certification (although the modules are likely such rated), I personally would not be concerned about using this amplifier in my own system.

-----
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
I just reviewed the 6-channel Buckeye Hypex class D amplifier. This is a follow up teardown to that review.

The amplifier is an assembly affair as all active electronics are contained in the Hypex NC252MP stereo class D amplifiers with integrated switching power supplies:

View attachment 100587

So the main job is routing the wiring to each module and input/output terminals. I was pleasantly surprised by the quality of the construction. While I will comment on some areas of improvement, they are really minor.

AC power input is from top right through the fuse and routed using the twisted read and white wires. While this coloring scheme is consistent with DC wiring, it should not be used for AC wires. Otherwise someone working on them may confuse them with DC low voltage wires with catastrophic consequences. The convention when AC and DC circuits are present in the same device as is the case here, is to use black or brown for "hot" and white or blue for neutral. The grounding terminal likewise uses a black wire and should instead use solid green or green and yellow wire:

View attachment 100588

I was pleased to see that the anodization was sanded off where the screw terminal is for safety ground. An ordinary washer is used. For belts and suspenders purpose, it is better to use a barbed washer that has "teeth" that really dig into the metal.

Plastic double sided taped clamps hold the wires (e.g. mains). These come loose over time and should be avoided. There are screws near all the wires so I suggest using a screwed in clamp. That said, I can't see anything bad happening if they come loose as they are not going to touch any sharp/hot metal.

Throughout the unit I found fine aluminum shavings/bits. Many were tiny but some were large enough to dislodge and cause a short in the fine pitch surface mount PC boards. You can see the specs in my larger shot above on the right near the speaker/XLR terminals.

Speaking of those terminals, they are all screw mounted which rules out me critiquing the soldering job. :) All were tight and tidy:

View attachment 100589

Using adhesive lined shrink tube is a very nice touch and above norm for such builds or even commercial ones. The adhesive keeps air out and should keep the tubing in place past my lifetime probably.

Zooming in a bit into Hypex modules we see the same practice of using second or third tier electrolytic capacitors:

View attachment 100590

There are large vents on top that should keep the ambient temps low inside the unit and help with longevity of the unit. Note however that the main source of heat are hidden power transistors that couple to the bottom of the chassis. Don't put the amplifier on top of any other heat generating component. I can't tell if there are heatsink compounds between the amps and the bottom plate. If not, some should be there.

Conclusions
I usually expect the worst when an audiophile decides to build such amplifiers before the teardown. I was super happy to see that some of the best safety and signal integrity rules have been followed here leaving me with some minor nits. While the overall unit has no safety or regulatory certification (although the modules are likely such rated), I personally would not be concerned about using this amplifier in my own system.

-----
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/


Hello to ALL: How do buy one of these AMP Units? How does one go about buying a Hypex NC252MP core? Or even a premed one? Cheers
 
I'd like to see more DIY'ers locating mains and signal on opposite ends of the chassis. (This is preferable in a number of ways.)
This is my twelve-channel prototype from a few years ago.

Dave.
 

Attachments

  • P1000820.jpg
    P1000820.jpg
    226 KB · Views: 1,054
A few more thoughts.
First up, wire colour - yes - please use the proper colours. (Being in Oz, we follow the EU colours, brown and blue for active and neutral, and stripped yellow/green for safety ground. I really don't like other standards, as black and white get used for other things.) No matter what you go with, don't use the colours for anything else.

You are using red and black inconsistently. You really need even more colours. The standby output supply is taken on a black wire out, and returns from the relay as red. Don't use black either. It gets generally recognised as negative or zero volts. Here it is +5v.

I’m afraid this not entirely correct, see https://www.electricaltechnology.org/2020/07/electrical-wiring-color-codes-nec-iec.html. If Australia would follow EU-code, in single phase 240V black should only be used for live-wire behind a switch, NOT for negative or zero. However, according to the website I linked to, Australia and New-Zealand have their own code... just like Russia, China, Japan, India & Pakistan, South-Africa, Israël, Canada, the U.S. all have their own... :facepalm:
So, good luck to Buckeye and any other manufacturer to comply with each and every rule. :p
 
Last edited:
I’m afraid this not entirely correct, see https://www.electricaltechnology.org/2020/07/electrical-wiring-color-codes-nec-iec.html. If Australia would follow EU-code, in single phase 240V black should only be used for live-wire behind a switch, NOT for negative or zero. However, according to the website I linked to, Australia and New-Zealand have their own code... just like Russia, China, Japan, India & Pakistan, South-Africa, Israël, Canada, the U.S. all have their own... :facepalm:
So, good luck to Buckeye and any other manufacturer to comply with each and every rule. :p
Those standards are for cabling in buildings. You can use any colour you want inside an equipment as long as it says "There are no user serviceable parts inside". The cables then becomes part of the equipment and not the installation, where the standards apply. However, as this is a hobby project, it is "helpful" to differentiate the high voltage AC power lines from the signal lines, not to mention to isolate them from each other for safety.
 
Last edited:
Back
Top Bottom