Hypex Ncore

Status
Not open for further replies.
I just made a rca->xlr connector per the diagram in the ncore datasheet and tried the rca out of my dac instead of the xlr, output voltage is even higher then xlr->xlr. Guess it's cap and resistor time?

Just out of curiosity I went back and measured the DC offset of all 8 nc400's I have (from the two first batches): 28, 26, 14, 6, 17, 21, 9 and 12 mV respectively (average values after letting the voltage settle for 5 min or so). Nicely within specs. With an unbalanced cable connected, but without the suggested 47K resistor fix, the offsets went up significantly.
 
I have read those documents. There in no mention of balanced bridged operation. Nor is there a sonic descripton of running balanced versus unbalanced. I am asking if the amps "sound better" in balanced mode and if they can be run in balanced bridged mode. Anyone?

I can't imagine how you could run them fully balanced. The output is in series with a single inductor, pairing another inductor would make the circuit unstable, I think. Also the general topology is balanced in the pre-output stages anyway. I expect that balancing this already-mostly-balanced circuit would gain you nothing except noise. Understand that both positive and negative parts of the signal are modulated by the oscillator, which would have to feed each half and so lose independence at that point.

Re bridging, I see no advantage there...you do that to trade current for voltage output, and it's a waste of the quality of the circuit, a quest for squeezing the most SPL out of an overstressed PA system, IMO. I haven't bothered to consider whether it's any more feasible than balancing. But I would welcome a contrary view, I haven't seen those circuits in a long time...
 
Julf & Mark... you guys are saying that instead of caps I should try one resistor across each xlr input across the hot/cold? Would the 47k resistors I tried earlier work?

Also if I put these resistors or use caps, will these have any affect on the sound quality/sonics of the amp?

Robbbby, I haven't addressed this yet because I thought it would be solved by now. First, your DAC has too much DC on the output, there is no excuse for 25mV. The pops you experienced are big because of that as well as transients that happen on turn-off, but those should be greatly attenuated by the DAC's (um, *missing*) protection circuitry. The bit about the amp needing an input cap is bull: this is a REFERENCE QUALITY DEVICE and its performance should not be compromised by an input cap. Want one? Have the DAC builder send you a teflon cap ($$$$) and bulk-foil resistor($$) gratis and call it good. Checking the amp is a snap, just run it with + and - inputs connected through some low resistance like 100 Ohms, and then connect those in series with a resistance above 1k to ground. You could probably tie them all together, but a little resistance never hurts. Run the output into a resistive load such as an 8-Ohm 50W resistor and watch the voltage, both AC and DC, and any peaks on your meter. Then plug in your amp. If outputs are very low, you're amp is fine. And 0.5V won't hurt your speaker, but it distorts the output a little---probably not enough to hear, but why tolerate it? BTW, if the amp plays a couple seconds after you unpower it, it has a sufficiently energetic power supply. Sure, muting is a nice feature when the amp is off, but not needed. We made amps that played for 2 minutes after unplugging, because we had half a farad of filter capacitance in the PS....SMPS perform as well with 1/100th the capacitance, so are vastly greener!

Try some other sources to compare: test DC on a cheap CD player or receiver or preamp output when there is no music (no CD playing, or the device is muted). Anything over 10mV is rare. Less than 5mV is typical.
 
Last edited:
your DAC has too much DC on the output, there is no excuse for 25mV.
I agree.

this is a REFERENCE QUALITY DEVICE and its performance should not be compromised by an input cap.
Not sure a pair of normal, decent poly caps ($1.50) would compromise the performance.

Checking the amp is a snap, just run it with + and - inputs connected through some low resistance like 100 Ohms, and then connect those in series with a resistance above 1k to ground.
Why? Why not just connect + and - inputs together? And definitely don't connect anything to ground. Then just measure the DC on the output.
 
Ok I will ! Funny this I have built over 100 diy project in my time ! :( Anyway I do not have pin 1 this is RCA input.

Maybe you should show these guys some of your finest creations. This could give some kind of indication on where the problem is located. In the Hypex modules or the actual builder...:

An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.
 
Not sure what you mean by "balanced mode". The amp always runs in "balanced mode" in that it is a fully differential amplifier. Are you talking about connecting it to a balanced vs. unbalanced source? I have not heard any audible difference (apart from some hum being picked up by some sloppy cabling when I was testing it with an unbalanced source), but DC offset definitely seems to increase with an unbalanced cable.

I am sure he mean one amp for each phase.
 
Thanks for your input it is much appreciated. Like already mentioned the designer of the dac said it is totally normal and that lots of reference quality source equipment has that kind of small amount of DC. The tech there checked the dac on the bench and it actually had 29mV, which is slightly higher than mine. No sense in trying to argue anything with them or trying to get them to fix it since I am in no position to debate anything electronics related.

I've been talking to another knowledgeable person just to get opinions on what I should do and he also suggested caps. He told me I should try capacitors only to see if the input impedance would work as the resistive part of the filter before adding resistors. He suggested 1uf caps with the 104k ohm input impedance since that would give me a 1.6Hz cutoff frequency. He also said it is trickier with the balanced inputs and that I definitely don't want to unbalance them so the 2x caps per channel I use need to match exactly and be with 0.1% or better of each other.

Since I know the 500mV of DC probably wouldn't really hurt anything when turning on the speakers and that i'll probably never hear the extra distortion, i'm taking my time and trying to figure out the 100% best way to do this so I don't have to do it twice. Anyone have any input on the numbers I just mentioned above?


Robbbby, I haven't addressed this yet because I thought it would be solved by now. First, your DAC has too much DC on the output, there is no excuse for 25mV. The pops you experienced are big because of that as well as transients that happen on turn-off, but those should be greatly attenuated by the DAC's (um, *missing*) protection circuitry. The bit about the amp needing an input cap is bull: this is a REFERENCE QUALITY DEVICE and its performance should not be compromised by an input cap. Want one? Have the DAC builder send you a teflon cap ($$$$) and bulk-foil resistor($$) gratis and call it good. Checking the amp is a snap, just run it with + and - inputs connected through some low resistance like 100 Ohms, and then connect those in series with a resistance above 1k to ground. You could probably tie them all together, but a little resistance never hurts. Run the output into a resistive load such as an 8-Ohm 50W resistor and watch the voltage, both AC and DC, and any peaks on your meter. Then plug in your amp. If outputs are very low, you're amp is fine. And 0.5V won't hurt your speaker, but it distorts the output a little---probably not enough to hear, but why tolerate it? BTW, if the amp plays a couple seconds after you unpower it, it has a sufficiently energetic power supply. Sure, muting is a nice feature when the amp is off, but not needed. We made amps that played for 2 minutes after unplugging, because we had half a farad of filter capacitance in the PS....SMPS perform as well with 1/100th the capacitance, so are vastly greener!

Try some other sources to compare: test DC on a cheap CD player or receiver or preamp output when there is no music (no CD playing, or the device is muted). Anything over 10mV is rare. Less than 5mV is typical.
 
Last edited:
Thanks for your input it is much appreciated. Like already mentioned the designer of the dac said it is totally normal and that lots of reference quality source equipment has that kind of small amount of DC. The tech there checked the dac on the bench and it actually had 29mV, which is slightly higher than mine.

I've been talking to another knowledgeable person just to get opinions on what I should do and he also suggested caps. He told me I should try capacitors only to see if the input impedance would work as the resistive part of the filter before adding resistors. He suggested 1uf caps with the 104k ohm input impedance since that would give me a 1.6Hz cutoff frequency. He also said it is trickier with the balanced inputs and that I definitely don't want to unbalance them so the 2x caps per channel I use need to match exactly and be with 0.1% or better of each other.

Since I know the 500mV of DC probably wouldn't really hurt anything when turning on the speakers and that i'll probably never hear the extra distortion, i'm taking my time and trying to figure out the 100% best way to do this so I don't have to do it twice. Anyone have any input on the numbers I just mentioned above?

Come on do you really think it matters if the cutoff is 1.6 hz in one phase and 1,584 hz in the other ? What a crappy tech ! :D Resistors would be something else.

Anyway my friend who had the NC during some days will not buy this amps. They are fantastic in many ways but the music is not free of the speakers he said.
 
Maybe if he had the intended and properly working power supply for the amps he would think differently

Well I have heard many Class D with " Proper " supplys ! In that case the situation with sound from a little " box " was even worse . But to be honest I can not tell if Brunos supply is much better.

In fact I have a little hum. But I did not have time to further exsperience since it was " finnished " half an hour before it went out of my place.

I am working on it now.

Do you know Niels Larsen from Denmark. He made the first Primare 928 and also Some Gryphon amps. I modded a NLE 1000 he made . It is a 500 W amp.
Really good work he said. SO it can not be that bad !
 
No sense in trying to argue anything with them or trying to get them to fix it

Fair enough :)

He told me I should try capacitors only to see if the input impedance would work as the resistive part of the filter before adding resistors. He suggested 1uf caps with the 104k ohm input impedance since that would give me a 1.6Hz cutoff frequency.

That would work just fine for getting rid of the DC on the speaker output - the resistors would be to try to lower the energy of the transient if you turn the DAC on or off with the amp on. 1uF is fine, but you would be OK with even 0.1uF - giving a cutoff of 16 Hz.

I suggest a pair of polycarbonate or polypropylene caps (25V is more than enough) per amp (or go teflon/PTFE if you want to be fancy).

He also said it is trickier with the balanced inputs and that I definitely don't want to unbalance them so the 2x caps per channel I use need to match exactly and be with 0.1% or better of each other.

Not sure I agree with that. There is a lot of superstition about balanced just because it is less common. Somehow people think in terms of signals with reference to the ground, when they should understand that the input of the amp only cares about the voltage between the + and - inputs, and doesn't care about ground at all.
 
Well I have heard many Class D with " Proper " supplys ! In that case the situation with sound from a little " box " was even worse . But to be honest I can not tell if Brunos supply is much better.

In fact I have a little hum. But I did not have time to further exsperience since it was " finnished " half an hour before it went out of my place.

I am working on it now.

Do you know Niels Larsen from Denmark. He made the first Primare 928 and also Some Gryphon amps. I modded a NLE 1000 he made . It is a 500 W amp.
Really good work he said. SO it can not be that bad !

I was just messing around with you, I had to go back and edit my post to put a smiley face just so you knew I was joking around and having some fun.
 
You never answered my question about your grounding strategy - somehow I got the impression that you might have gone for a star topology?

No its more a bus than star I would say. But the gnd from the power supply to the input section must be connected somewhere. Earlier I asked if it could be a idea to use one of the four gnd leads from the board to connect the " little supply ". ???
 
No its more a bus than star I would say.

OK, so where does signal ground connect to chassis/power ground?

But the gnd from the power supply to the input section must be connected somewhere. Earlier I asked if it could be a idea to use one of the four gnd leads from the board to connect the " little supply ". ???

We are talking about Vsig, the symmetric supply to the op amps? It's ground connects to the same ground as the main power supply, pins 3,4,9 and 10 on J7. I assume you are not connecting the optional driver supply voltage (Vdr, pin 11) at all?
 
Status
Not open for further replies.