[HighRez]

Hello,

I am in the process of planning my NCore build and have a few questions I was hoping someone folks can answer regarding aspects of the power cabling that is involved and supplied with the unit.

Power Input:

Question # 1: If I was intending to build a Dual Mono amp with 2 x SMPS600 & 2 x NC400 modules in the same chassis is there any reason to believe that using only one IEC Input to feed both SMPS600's is a weak point verses using two IEC Inputs (ie..one per SMPS600). Are there any reasons why using only one IEC Input is better than two short of aesthetics? The only thing I could come up with is that maybe using two separate IEC Inputs would introduce some kind of ground loop but I'm not sure.

Question # 2: Would it be a good or bad idea to use 2 X brass/copper power distribution blocks to terminate the input wires attached to the IEC Input connector and then from the distribution block attach the included cables to feed the SMPS600 power supplies? My thoughts on using this method would be that it may help ensure that each SMPS600 power supply draws power equally from the feed as to avoid the possibility of one SMPS600 "seeing" the power first and taking more then its neighbor.

Question # 3: Would terminating the IEC wall feed wires first to a distribution block within the chassis before feeding the SMPS600 generate more noise that may not be there if they were not used?

Question # 4: Is it safe to assume that keeping the power wires feeding the SMPS600 as short as possible is better then using the default lengths supplied? If keeping these wires as short as possible is considered ideal then would that be reason not to use a distribution block method due to the added wiring which would be required? Is adding 6 more inches of wire the end of the world considering all the other wire involved upstream of the SMPS600 starting at the breaker box, thru the AC wall outlet, thru the power cable to the IEC connector?

Thanks for your thoughts and time on these matters

See Diagram below for an example of what I mean for questions 2,3 above:

[HighRez]

Figured I should add a diagram of my design so far to put things into better perspective as to why the SMPS600's are layed out the way they are and why I am considering the distribution blocks. 

[mjosef]

If it was me, I would have the IEC located in the middle of the backplate, make a Y wire harness with equal lengths for the two feed-thru to the two power supplies. A simple copper crimp at the 'neck' of the Y would be the 'joint' for the three wires.

[Julf]

If it was me, I would have the IEC located in the middle of the backplate, make a Y wire harness with equal lengths for the two feed-thru to the two power supplies. A simple copper crimp at the 'neck' of the Y would be the 'joint' for the three wires.

Agree. Keep it simple and short.

Now, for the next debate - should the SMPS600's be wired in phase (J1 pin 1 to hot, pin 3 to neutral on both SMPS's), or anti-phase (one opposed to the other)?

[gstew]

And this is why:

One thing I found was that connecting the hot to pin 2 and the ground to pin 1 (closest to the mounting hole) on the SMPS's produced the lowest stray AC on the unit's ground. This was consistent across the three SMPS600's I tested.

With past amplifier builds using linear supplies, orienting the AC input this way produced noticably better sound (no Julf, no double-blind tests were performed in my developing this understanding, so don't bother to ask!). I did not test whether reversing the AC input connections made a difference on the NCores, but from my past experience, orienting the AC the way that produces the lowest stray voltage is always the best for any power supply, amplifier, preamp, or other.

To test this with a linear supply,  I mount the transformer to the chassis, wired to the AC via a fuse, but unconnected to anything else with the output leads/connections safely insulated. I then use a cheater plug to plug the cord in with the AC prongs one way, then the other, and measure from the chassis to the ground on the AC plug using the lowest AC setting of my meter.

This is easier to measure on an I-E transformer where the mounting lugs are electrically connected to the laminations, but is still measurable on a toriod with a metal mounting bolt through the center. I find this is easier to discern with an analog meter, but can be done with a digital one. In some cases, the difference is very small, a volt or two. In others, it can be as much as 20-30 VAC. My SMPS's were in the latter camp.

When I did this with the SMPS600's, I made VERY sure that the DC outputs had drained down to 0v before connecting the NC400 modules by letting them set overnight AND then measuring. If they have significant DC still in the output caps, that may damage either the SMPS600 or the NC400 when you connect them. BE WARNED!

Greg in Mississippi

[HighRez]

Thanks for the replies.

The suggestion to use a copper crimp instead of the distribution blocks I have shown is appreciated and sounds like a good alternative that I hadn't thought of. I like that the crimp would have less mass and thus be less likely to act as an antenna for noise while providing the same "even" flow/distribution of electricity to the SMPS modules.

As you can probably tell I have purposely decided to go against the grain in terms of where to place the Neutrik PowerCon Input i intend to use on the chassis. Despite this location not being as "pretty" as having it on the back of the chassis I feel that keeping those wires as isolated as possible thru whatever means necessary is a more ideal way to go. Will it make a difference? I'll probably never know but I will sleep better believing that I did what I could to keep that noisy wire away from everything else...   

More of the same can be said about isolating the SMPS modules into their own chambers away from each other and away from the NC400 modules. Will it make a difference? Again, I'll probably never know. Is it overkill? Absolutely!

Hello GStew,

Not sure I understand your reply. Are you saying you would do the same as I am intending to do in my diagram or are you are saying you would do the same as Julf mentioned about wiring "In Phase/Out Phase"? Anyone care to post a quick and dirty diagram of this wiring method so a poor slob like myself better understands his options? 


Thanks again.

[gstew]

Sorry for not being clear.

Wire the AC input so that the AC hot goes to pin 2 on both SMPS's and the AC neutral to pin 1 (closest to the mounting hole) on both SMPS's.

Your diagram has the hot going to pin 1 on one of the SMPS and pin 2 on the other. You want the AC input to be wired the same on both.

Greg in Mississippi

[Julf]

One thing I found was that connecting the hot to pin 2 and the ground to pin 1 (closest to the mounting hole) on the SMPS's produced the lowest stray AC on the unit's ground. This was consistent across the three SMPS600's I tested.

That of course only works if you have a 3-pin mains connector, or a "polarized", unsymmetrical 2-pin connector. Doesn't really work with continental European power plugs that don't have polarity.

I am a bit confused about your measurements - fully understand how you measure it from a transformer with a linear supply, but not quite sure how you measured it on the SMPS - from where did you measure the AC differential to ground?

[gstew]

That of course only works if you have a 3-pin mains connector, or a "polarized", unsymmetrical 2-pin connector. Doesn't really work with continental European power plugs that don't have polarity.

I really don't know how European power is setup. Is it balanced or ground referenced?

I am a bit confused about your measurements - fully understand how you measure it from a transformer with a linear supply, but not quite sure how you measured it on the SMPS - from where did you measure the AC differential to ground?

I just mounted the SMPS into the chassis & applied power. Since it is grounded to the chassis through one of the standoffs, any stray voltage on the ground also showed up on the chassis. I measured from the chassis to the AC outlet ground.

The result was interesting to me. I had never scratch-built a unit with an SMPS before & never tried to measure induced voltage on this type of supply. But after thinking about it (I have a fairly basic understanding of SMPS operation, not much more), it did make sense that it could just as easily have a different induced voltage based on AC input polarity.

I have several amps that use B&O IcePower modules with integrated SMPS's... I need to check them too!

Greg in Mississippi

[Julf]

I really don't know how European power is setup. Is it balanced or ground referenced?

Balanced. There are two different kinds of power plugs and sockets (just like in the US). "Normal" ones are 2-prong and can be inserted either way, and there are 2 different 3-pole "safety" ("schuko") plugs/outlets that have safety earth as well, one type (the Euro standard)  allows insertion either way, the other (in use in France, Belgium and Poland) has a offset third prong, so it doesn't.

I just mounted the SMPS into the chassis & applied power. Since it is grounded to the chassis through one of the standoffs, any stray voltage on the ground also showed up on the chassis. I measured from the chassis to the AC outlet ground.

Right. If both the source and the amp follows the Hypex recommendation of not connecting mains ground to the chassis (or the connection between the source and the amp is balanced without ground/earth connection) it shouldn't matter, but if there are ground connections, it could indeed make a difference with a ground-referenced mains. 

[Occam]

I really don't know how European power is setup. Is it balanced or ground referenced?
........

Balanced. .....

North American 240vac is balanced (120vac-0-120vac), split phase. European mains 240vac  is single ended with a 240vac hot, and a grounded neutral. Both are ground referenced, i.e. not 'isolated power' as found in hospitals. FWIW

PS - I do believe Julf knows this, but he is not posting in his native language. Props to him for his eloquence in English.

[Julf]

Ah, yes, thank you for the correction, Occam (but careful with the razor!).

As I was posting, I was actually going "now, did I get the terminology right".

European houses usually have a 3-phase (4 cables) 400 V connection. Inside the building, each phase is either split off to power individual circuits at 230 V, by referencing one of the 3 phases to neutral, or used to power higher-power-consumption devices such as ovens and heating boilers at full 3-phase 400V. So yes, the single-phase 230V is referenced to neutral/ground.

[Julf]

North American 240vac is balanced (120vac-0-120vac), split phase.

Just checking - not that familiar with the US system, but aren't individual outlets single-ended, neutral-referenced 120V circuits?

[Occam]

Just checking - not that familiar with the US system, but aren't individual outlets single-ended, neutral-referenced 120V circuits?

Julf,

Yes, most mains powered appliances in North America is run off of 120vac hot and a grounded neutral. As 3 phase is not common in residences, we have what 'old folks' (me) called 'house power' entering our mains panel, which are 2 anti-phase 120vac lines, from either of which we can wire 120vac circuits, but for large appliances we derive 240vac from both phases, balanced.

Regards,
Paul

PS - Here is a reference to the technique Gstew mentioned regarding minimization of chassis leakage -
http://www.boundforsound.com/tweak.htm
My only caveat is that one can reverse the wiring of a transformer's primary or smps, to minimize to chassis leakage, but ONLY after an on-off switch and fuse. If you reverse hot and neutral on a cord or IEC inlet, you compromise the safety functionality of the switch and fuse!

[Julf]

Yes, most mains powered appliances in North America is run off of 120vac hot and a grounded neutral. As 3 phase is not common in residences, we have what 'old folks' (me) called 'house power' entering our mains panel, which are 2 anti-phase 120vac lines, from either of which we can wire 120vac circuits, but for large appliances we derive 240vac from both phases, balanced.

Paul,

Right, so sounds like a 2-phase version of the European 3-phase system - instead of 2 anti-phase lines, we have 3 120 degree phase lines. Each of the 3 can be used referenced to earth/ground for 230 V, but larger appliances you can use all 3 phases, with or without referencing to earth/ground.

Still, going back to the original issue - standard European 230V single-phase plugs and sockets are unpolarized, so you can insert the plug either way. The larger 3-phase plugs and sockets are of course keyed to go only in one position.

My only caveat is that one can reverse the wiring of a transformer's primary or smps, to minimize to chassis leakage, but ONLY after an on-off switch and fuse. If you reverse hot and neutral on a cord or IEC inlet, you compromise the safety functionality of the switch and fuse!

Learned that 35 years ago changing a glass fuse. 230 V gives you a much more significant jolt than 120 V
 
I guess growing up in a world of "either pole can be hot", I have pretty much always used double-pole switches.

Cheers and thanks,

       Julf