Agree with Julf on heat dissipation. Even Bruno has suggested that heat sink compound is not neccessary between NC-400 module and the mounting surface because of the large surface area of contact, and removing anodization from such a large area would be tedious at best! Just make sure the surfaces are clean of any debris, and the module heat sink bolts down tightly, the heat transfer should be entirely adequate then. The NC-400 heatsinks, even resting on a piece of wood and being being pushed pretty hard, do not get all that hot.
NC-400 modules are grounded via the shield to chassis connection of their cables.
NC-400 modules are grounded via the shield to chassis connection of their cables.
IEC w/ Fuse?
I am in the process of trying my hand at learning/building an ncore amp, and had a question on the IEC. If i don't use a switch, will i need an IEC with a built in fuse? i know the PSU has a built in fuse, but without a switch, wont the chassis be exposed without a fuse/switch at the IEC? thanks george
I am in the process of trying my hand at learning/building an ncore amp, and had a question on the IEC. If i don't use a switch, will i need an IEC with a built in fuse? i know the PSU has a built in fuse, but without a switch, wont the chassis be exposed without a fuse/switch at the IEC? thanks george
Not sure what you mean by "exposed". The power supply fuse will protect anything downstream from the power supply, and if you get a short circuit before the power supply, it will trigger your house mains circuit breaker.
I...
did my stereo NC-400 with 2xSMPS 600 amp with a Furutech IEC with no fuse or switch. I just made absolutely sure that the input power wiring is totally secure and could not come loose and short to the chassis. The wiring is well soldered at both the IEC and SMPS 600 ends, the wire is very well insulted with teflon, is secured inside the chassis to where it never rubs on anything (in case I were to ship the amp and it was exposed to hours of vibration) and is at least 1/2" from any chassis part or other component.
If you are not absolutely certain your wiring is secure and cannot cause a short, then an IEC with fuse is the best insurance.
did my stereo NC-400 with 2xSMPS 600 amp with a Furutech IEC with no fuse or switch. I just made absolutely sure that the input power wiring is totally secure and could not come loose and short to the chassis. The wiring is well soldered at both the IEC and SMPS 600 ends, the wire is very well insulted with teflon, is secured inside the chassis to where it never rubs on anything (in case I were to ship the amp and it was exposed to hours of vibration) and is at least 1/2" from any chassis part or other component.
If you are not absolutely certain your wiring is secure and cannot cause a short, then an IEC with fuse is the best insurance.
If your chassis is safety grounded, and something shorts, either the mains circuit breaker or the ground fault interrupter (RCD or RCBO) will trip.
If, on the other hand, your chassis isn't safety grounded, and you worry about the security of your wiring, you have created a death trap and no circuit breaker or fuse in the world will save you.
Some experience of residual noise cancelling
I’ve just built a new stereo amp using the Ghent Audio Stereo Case Kit and SMPS1200A400.
The amp is connected to Denon AVR-X4000 pre-amp output (unbalanced RCA). For connection are used two Choseal 4N-OFC RCA(M) to XLR(M) Cables 1m. The cables are of good quality according to nc400 datasheet recommendations.
However connecting the amp I’ve issued noise coming from the residual current. The problem is that the cable shield connects both amp and pre-amp chassis. It can be easy shown using the Figure 3 from the nc400 datasheet:
Red dotted line shows the way of the residual current.
This problem I’ve corrected by following modification:
1. In the cable I’ve cutted the shield connection on one side. In my case it was easier to cut on the XLR side, but I’d recommend to cut on the RCA (left) side, because in such case the shield is better isolated from the signal wires.
2. Put in the back plate of the Chent case additional terminal for case ground connection and connect both devices by any cable:
Residual current, if any, flows no more via the shield but via the external dedicated cable generating no noise in audio lines.
In my case after the modification the noise is completely disappeared.
-----------------------------------------------------
my system: AVR-X4000 + 2x nc400 -> 2x Harwood Reference (Hobby HiFi 2/14) + 5x Lancetta (Hobby HiFi 6/04)
I’ve just built a new stereo amp using the Ghent Audio Stereo Case Kit and SMPS1200A400.
The amp is connected to Denon AVR-X4000 pre-amp output (unbalanced RCA). For connection are used two Choseal 4N-OFC RCA(M) to XLR(M) Cables 1m. The cables are of good quality according to nc400 datasheet recommendations.
However connecting the amp I’ve issued noise coming from the residual current. The problem is that the cable shield connects both amp and pre-amp chassis. It can be easy shown using the Figure 3 from the nc400 datasheet:
An externally hosted image should be here but it was not working when we last tested it.
Red dotted line shows the way of the residual current.
This problem I’ve corrected by following modification:
1. In the cable I’ve cutted the shield connection on one side. In my case it was easier to cut on the XLR side, but I’d recommend to cut on the RCA (left) side, because in such case the shield is better isolated from the signal wires.
2. Put in the back plate of the Chent case additional terminal for case ground connection and connect both devices by any cable:
An externally hosted image should be here but it was not working when we last tested it.
Residual current, if any, flows no more via the shield but via the external dedicated cable generating no noise in audio lines.
In my case after the modification the noise is completely disappeared.
-----------------------------------------------------
my system: AVR-X4000 + 2x nc400 -> 2x Harwood Reference (Hobby HiFi 2/14) + 5x Lancetta (Hobby HiFi 6/04)
Hi,
I was checking DC offsets in my system (still trying to hunting down that stuttering noise when I connect NC400 to TP Buffalo DAC), and thus was measuring DC voltage on the speaker outputs and the balanced inputs.
On output side, everything is looking perfect, without anything connected and ampon grounded, the output dc I measured was 0mv (slowly decreasing to this value as warming up).
However, when I measure the amp input, I get some wild values of 250mV between ground and either non-inverting or inverting input. I measured this with ampon disconnected, no inputs or outputs connected, using the hypex supplied standard input cable (not connected to xlr inputs or anything, just used this to make it easier to connect the multimeter). There, I measured between the shield (=gnd) and both blue and white leads. I got values wandering between 200-280mV, sometimes even lower, even after warm-up.
I'm really wondering if I making some mistake here or is this really so high or is this ok like that?
Would be grateful for any advice.
Thanks,
igy
I was checking DC offsets in my system (still trying to hunting down that stuttering noise when I connect NC400 to TP Buffalo DAC), and thus was measuring DC voltage on the speaker outputs and the balanced inputs.
On output side, everything is looking perfect, without anything connected and ampon grounded, the output dc I measured was 0mv (slowly decreasing to this value as warming up).
However, when I measure the amp input, I get some wild values of 250mV between ground and either non-inverting or inverting input. I measured this with ampon disconnected, no inputs or outputs connected, using the hypex supplied standard input cable (not connected to xlr inputs or anything, just used this to make it easier to connect the multimeter). There, I measured between the shield (=gnd) and both blue and white leads. I got values wandering between 200-280mV, sometimes even lower, even after warm-up.
I'm really wondering if I making some mistake here or is this really so high or is this ok like that?
Would be grateful for any advice.
Thanks,
igy
It really doesn't matter - it is a differential input, so it is not referenced to ground - as proven by the fact that you don't have any DC on the output.
@Julf: in fact not. Later I've thought I could try, but was already done.
If shield remains connected it still can drive part of current depending on resistance (active & inductive) relative to the ground wire. And shields are usually made of bigger diameter (=coax, ~5mm in my cable) and can be a good antenna.
Cutting shield eliminates this risk.
If shield remains connected it still can drive part of current depending on resistance (active & inductive) relative to the ground wire. And shields are usually made of bigger diameter (=coax, ~5mm in my cable) and can be a good antenna.
Cutting shield eliminates this risk.
Still trying to identify source of stuttering noise
Hi All,
Since weeks I'm trying to identify the root cause of the stuttering noise at the output of NC400, when connected to my TP Buffalo III dac.
After spening quite some time analysing and filtering the attached file, I had the concolusion, that somehow it looks like if for some reason the noise would be amplified periodically 6-9dB more. More exactly, every 46ms there's a 14ms part where the output is simply louder, while keeping the same spectral structure. Now, 46ms period is roughly 22Hz.
Interestingly, the dac output has a spike just at this frequency, but it's extremely low in level, around -125dB at standard +4dBu line level, and I can see/hear no trace of amplitude changes like at the amp output. I'm wondering if this is a coincidence, or if this somehow makes the amp behaving strangely?
Things I ruled out so far:
- DC offset: tried adjusting both on the amp and using caps at the dac output. Make no difference at all
- Source: the dac behaves the same with my real dsp source and with an fpga board supplying arbitrary bitclock
- Running dac without output stage in voltage mode gives the very same results
- Single amp problem: 6 NC400 and 4 SMPS600 power supplies tested, producing the same result
Any hints what to try next?
Thanks in advance,
igy
Hi All,
Since weeks I'm trying to identify the root cause of the stuttering noise at the output of NC400, when connected to my TP Buffalo III dac.
After spening quite some time analysing and filtering the attached file, I had the concolusion, that somehow it looks like if for some reason the noise would be amplified periodically 6-9dB more. More exactly, every 46ms there's a 14ms part where the output is simply louder, while keeping the same spectral structure. Now, 46ms period is roughly 22Hz.
Interestingly, the dac output has a spike just at this frequency, but it's extremely low in level, around -125dB at standard +4dBu line level, and I can see/hear no trace of amplitude changes like at the amp output. I'm wondering if this is a coincidence, or if this somehow makes the amp behaving strangely?
Things I ruled out so far:
- DC offset: tried adjusting both on the amp and using caps at the dac output. Make no difference at all
- Source: the dac behaves the same with my real dsp source and with an fpga board supplying arbitrary bitclock
- Running dac without output stage in voltage mode gives the very same results
- Single amp problem: 6 NC400 and 4 SMPS600 power supplies tested, producing the same result
Any hints what to try next?
Thanks in advance,
igy
Ok, but then why does it make sense to have a trimmer pot to adjust it?
The trimmers are there mainly to minimize *output* DC offset - and you of course don't want a *differential* input offset.
It does sound like the problem is with your DAC. Any chance the 46 ms cycle could be bursts of HF oscillation?
I can try to check the dac output with scope, so far I've not yet done this. Do you think HF bursts could cause something like this?
Last edited:
I think it is worth doing.
More like "let's check every possibility" - as there clearly is something cyclical going on, but the 44.1 kHz sound files you posted show only very low-level signals. Thus I think it would be good to make sure there isn't anything going on at higher frequencies (that would not show up in your sample files).
- Status
- Not open for further replies.