Hypex NC252MP Overheating?

[YouCanCallMeGeorge]

I'm wondering if my Hypex NC252MP amplifier is overheating in its Ghent Audio case.

I bought the amplifier early this year. I left the entire system powered up 24 hours a day, even when not in use. A couple of months ago I noticed that the amplifier case was always warm to the touch, even after hours without being used. (I would say the temperature was 105 to 110 deg F; my hand was never uncomfortable on the case after extended contact.) Perhaps the amplifier case had always run warm; I never checked. I started turning the system off when not in use.

Twice in the last month or so the amplifier has malfunctioned.

Symptoms:
1. Volume of the left speaker drops to about 5% of the right speaker.
2. The sound from the left speaker is 50% music and 50% static. The static sounds exactly like a tuner set to an FM frequency on which no station is broadcasting. I take this as an indication that the amplifier module is generating white noise.

The first time this happened I didn't know whether it was the amplifier, the upstream equipment, or the speaker. I spent some time checking connections and switching input sources to the pre-amp. After about 15 minutes I tried cycling power to the entire system. At power up, the system behaved normally.

Last night, during the second malfunction, I began removing and swapping components (including cables) to isolate the malfunctioning component. The one and only constant was the left channel of the amplifier. Nothing I did made the system return to normal sound. After some time—perhaps 20 minutes, give or take—the system did return to normal sound. I noted that the amplifier case had cooled off, down very close to palm-of-hand temperature. I put all the components and cables back in their original positions, and the system ran perfectly.

 

[pma]

Unfortunately, many assemblers (assembling companies) do not use a proper heatsinking of the OEM class D modules. This is the tax for the fact that almost anyone is doing such job, these days. Not enough qualification and lot of underestimation of proper engineering practices.

 

[YouCanCallMeGeorge]

Unfortunately, many assemblers (assembling companies) do not use a proper heatsinking of the OEM class D modules. This is the tax for the fact that almost anyone is doing such job, these days. Not enough qualification and lot of underestimation of proper engineering practices.

The builder of this unit is working with me on the issue. So far, I'm the first person to report an overheat. There is plenty of empty air space all around the Ghent case, so if there is an overheating problem, it is with the Ghent case itself. I'll remove the front cover for now and see what happens. (No children in the house; not a shock hazard.)

 

[peng]

If I know I will be using the amp for more than a couple of hours I would turn on my external fan because they do get quite warm even just idling. I also believe the issue is that the low cost ones don't have heat sinks as they rely on the casing to dissipate the heat.

 

[YouCanCallMeGeorge]

If I know I will be using the amp for more than a couple of hours I would turn on my external fan because they do get quite warm even just idling. I also believe the issue is that the low cost ones don't have heat sinks as they rely on the casing to dissipate the heat.

That's good to know, as it means that the amplifier itself is probably working normally.
I'll do something to improve heat transfer out of the box.

 

[Willem]

Why would you leave it on? It wastes electricity and hence money. Also, it is neither good for the environment, nor for the amplifier.

 

[Kevinfc]

“Why would you leave it on? It wastes electricity and hence money. Also, it is neither good for the environment, nor for the amplifier.”
Isn’t that one of the great advantages of class D amplifiers? The standby watts required for a switching power supply at idle are nearly nothing. I’m all for conservation and efficiency, but at some point enough is enough. Less than one watt per hour is minuscule.

 

[Willem]

I happen to be the son of an electrical engineer with a distinguished career in industrial and military electrical engineering. His lesson to me was to always turn off any gear not in use, because you never know when it will malfunction and may catch fire. I guess I am just a son who does as he was told to do on a subject that his father knew far more about. I agree that class D is far better, but even so: in this case, the amp got hot, or at least warm enough to potentially malfunction (if that is what it was).

 

[peng]

“Why would you leave it on? It wastes electricity and hence money. Also, it is neither good for the environment, nor for the amplifier.”
Isn’t that one of the great advantages of class D amplifiers? The standby watts required for a switching power supply at idle are nearly nothing. I’m all for conservation and efficiency, but at some point enough is enough. Less than one watt per hour is minuscule.

Fair enough, but are you sure you left the unit on standby when not in use? Depending on the design/build, it may not go into standby by itself. For example, Buckeye amp would not include a standby switch if ordered with the Ghent case, so I added the switch myself, though he was nice enough to drill the hole for the switch. It is not automatic so I do leave the power switch in the on position and use the standby switch instead to keep it cool when not in use.

 

[AudioJester]

My experience with hypex series -UcD and ncore - is its usually the power supply that goes first, and the initial signs are volume/output differences between channels. Now obviously the nc252mp is a bit different with the integrated power supply and buffer. The power supplies dont do well left on 24/7 (even with the standby mode) - I learnt this the hard way, probably user error on my behalf.
There is no easy fix, other than replacing the power supply or with the nc252mp replacing the entire module.

 

[peng]

My experience with hypex series -UcD and ncore - is its usually the power supply that goes first, and the initial signs are volume/output differences between channels. Now obviously the nc252mp is a bit different with the integrated power supply and buffer. The power supplies dont do well left on 24/7 (even with the standby mode) - I learnt this the hard way, probably user error on my behalf.
There is no easy fix, other than replacing the power supply or with the nc252mp replacing the entire module.

Really, I wonder why even standby mode would be an issue? The data sheet says 0.25 W consumption when in standby with no load. I also checked with tech support and the response was positive. The amp is actually cold to the touch when left in standby.

 

[YouCanCallMeGeorge]

I left the entire system on 24/7 because I have read that the power cycle can stress electronics. That's a rule of thumb, and not necessarily applicable to the components in this system—but a couple of dollars per month is much easier to handle than hundreds of dollars all at once to replace a component, so why not?

I see that I can put the amplifier in standby mode by opening the jumper between pins 3 and 9 on J6. It's easy enough to put a switch on the front panel and attenuate the warming while idle.

The malfunction is very likely due to overheating, given the observed behavior and the tiny holes in the top of the Ghent box. I will probably have two wide slots milled into the front cover, one near the bottom of the cover for air inlet, the other near the top for outlet. For that matter, I could drill two rows of holes along each side of the case and get as much air flow as I may need. (Of course, I would take the amp out of the case before drilling.)

For now I will listen to long sessions of loud music with the front cover off and see what happens.

 

[peng]

I see that I can put the amplifier in standby mode by opening the jumper between pins 3 and 9 on J6. It's easy enough to put a switch on the front panel and attenuate the warming while idle.

That's what I did, in standby mode the amp is cold to the touch. It also virtually eliminate the pops that could be otherwise be audible when turning power on or off.

 

[restorer-john]

@YouCanCallMeGeorge

Go have a look at this post:

Class D amp long term reliability

Inexpensive today? A little over ten years ago Yamaha produced their NS-1000 speakers for $1000 a pair, and sold many. Today their replacement for the NS-1000 is their NS-5000 for $15,000 a pair, which granted their replacement is quite higher performance, but it sure doesn't seem to me like...

www.audiosciencereview.com

Loss of one channel was the lead up to this failure.

 

[YouCanCallMeGeorge]

@YouCanCallMeGeorge

Go have a look at this post:

Class D amp long term reliability

Inexpensive today? A little over ten years ago Yamaha produced their NS-1000 speakers for $1000 a pair, and sold many. Today their replacement for the NS-1000 is their NS-5000 for $15,000 a pair, which granted their replacement is quite higher performance, but it sure doesn't seem to me like...

www.audiosciencereview.com

Loss of one channel was the lead up to this failure.

Well, that's not encouraging. Here is a shot of my board, as seen from the open front panel—obviously warped. I'll just have to wait and see.

 

[RickS]

I left the entire system on 24/7 because I have read that the power cycle can stress electronics. That's a rule of thumb, and not necessarily applicable to the components in this system—but a couple of dollars per month is much easier to handle than hundreds of dollars all at once to replace a component, so why not?

I see that I can put the amplifier in standby mode by opening the jumper between pins 3 and 9 on J6. It's easy enough to put a switch on the front panel and attenuate the warming while idle.

The malfunction is very likely due to overheating, given the observed behavior and the tiny holes in the top of the Ghent box. I will probably have two wide slots milled into the front cover, one near the bottom of the cover for air inlet, the other near the top for outlet. For that matter, I could drill two rows of holes along each side of the case and get as much air flow as I may need. (Of course, I would take the amp out of the case before drilling.)

For now I will listen to long sessions of loud music with the front cover off and see what happens.

How many screws are holding the amp module to the case? Are they tight?

Was there any thermal grease applied?

Btw, I have 2 Ghent NC252MPs that I built. I use for speaker design and have been left on overnight several times and do not get overheated.

If you have the top off, you can also measure the temp of the outputs. There are thermistors built into the amp. See the data sheet for details.

 

[Madjalapeno]

So @restorer-john - is there anything that us mere mortals can do to extend the life? Would it be worth slathering certain components with non conductive heat sink compound?

When I built my Neurochrome amps the instructions from Tom are very clear about making sure there is a good path to the heatsinks.

 

[YouCanCallMeGeorge]

How many screws are holding the amp module to the case? Are they tight?

Was there any thermal grease applied?

Btw, I have 2 Ghent NC252MPs that I built. I use for speaker design and have been left on overnight several times and do not get overheated.

If you have the top off, you can also measure the temp of the outputs. There are thermistors built into the amp. See the data sheet for details.

Looks like nine screws holding the Hypex assembly to the Ghent case. I have not checked tightness. Thermal grease was not applied at assembly. (That's something to ask the assembler about before ordering a new unit.)

The top of the case is very close to room temperature with the front cover removed, after two hours running at the maximum tolerable volume in this room. The air temperature in the case above the amplifier modules is just a few degrees warmer than ambient. I see the pins for monitoring temperature. My inclination right now is that with proper ventilation there won't be any problem with module temperature. I can revisit the issue if I continue to have malfunctions after opening up the case for air flow.

This is the case that holds the NC252MP: https://www.ghentaudio.com/kit/2180a-2.html
Is this the case that you are using?

Also, to clarify, the overheating occurs after an hour or two of playing music. The unexpectedly warm temperature at idle was the signal to me that overheating might be the cause of the malfunction.

 

[YouCanCallMeGeorge]

That's what I did, in standby mode the amp is cold to the touch. It also virtually eliminate the pops that could be otherwise be audible when turning power on or off.

Apparently those pops can damage the amplifier if the voltage spike is of sufficiently high voltage and duration. See question 7 here:https://www.hypex.nl/faq/ (The question refers to a tube pre-amp. My preamp is a solid state SHD unit, so not the same conditions. But the point remains that it is better to enable the power amp after the pre-amp has stabilized.)

 

[RickS]

Looks like nine screws holding the Hypex assembly to the Ghent case. I have not checked tightness. Thermal grease was not applied at assembly. (That's something to ask the assembler about before ordering a new unit.)

The top of the case is very close to room temperature with the front cover removed, after two hours running at the maximum tolerable volume in this room. The air temperature in the case above the amplifier modules is just a few degrees warmer than ambient. I see the pins for monitoring temperature. My inclination right now is that with proper ventilation there won't be any problem with module temperature. I can revisit the issue if I continue to have malfunctions after opening up the case for air flow.

This is the case that holds the NC252MP: https://www.ghentaudio.com/kit/2180a-2.html
Is this the case that you are using?

Also, to clarify, the overheating occurs after an hour or two of playing music. The unexpectedly warm temperature at idle was the signal to me that overheating might be the cause of the malfunction.

Yes, same case.