New amps from hypex

[TEK]

Hi

Thought I should mention that it seems like if Hypex is releasing a new series of OEM products:twisted:
NC500MP
NC502MP (2 channel)
From what I understand this is a one-in-all product where the power supply and amp module is all delivered in one package.

Datasheet

I have not looked in details about how this is compared to a SMPS1200 + NC500
A NC500MP seems to be around 140 Euro. A NC502MP is around 250 Euro.

If anyone feels like commenting on how a NC500MP would perform compared to a NC500 and a SMPS1200A700 power module - I'm all ears:E

[dar47]

Hum, seams like just an all in one package. I still think we are golden with the NC500 and SMPS1200. Maybe these are to widen the market to those who don't want to do there own thing. Jon would probably know if our planned path still has merit. These may be more compact for that 5 channel option though. I wouldn't be worried if you choose to get these new ones for the 5 channel as I am sure you could sale your stock no problem!

[JonMarsh]

Mike is planning on working with these, specifically the 502, if I recall correctly. (I'll check my emails later this week and get back to you- I've been awfully busy for it theoretically being a week off! )

The THD+N curves versus frequency look pretty nice, especially at low power (they should- at this operating point the amp is working in ZVS switching)l No two tone HF CCIF IM distortion plots; if anything, that might be their weak point (it is for pretty much everything short of a Halcro DMX8 series amp)

Were one doing a powered speaker, these would be a very logical choice. Interesting that they're going to offer the tweeter power amp option, too.

Is this series available to you also, TEK?

At work I'm going to be working on a SiC MOSFET based LLC design Telecom supply that could be easily adapted to Class D amplifier applications... the SMPS will run at a nominal 240kHz. Should be interesting... as long as my manager doesn't pull a Lucy on me. (think Charlie Brown and Lucy, and the football...)

[TEK]

Yes, they are available to me.
It's 140 euro for the nc500np and 250 for the nc502np ex vat.
Seems to be a evaluation kit as well for 25 euro.

https://www.diyclassd.com/documenten/download/920

(Not sure if you havevaccess to the datasheet)

[TEK]

Any thoughts about the differences between:
1) NC500 + SMPSA700 vs NC500MP
2) 2 NC500 + 2 SMPSA700 vs NC502MP

[JonMarsh]

That's one of those things that I would not offer an opinion unless I'd measured and listened...

Keeping in line with the idea that it's better to be quiet and thought a fool, than to speak forth and remove all doubt! :B

However, I do have to say that on paper, nothing seems to suffer from the integration in the NC500MP. In the power ranges that count, THD+N is very good across the band, you have to go to Halcro turf to significantly improve on it. The only other thing I wish they provided measurement data on is CCIF two tone HF IM distortion. I'm sure they know how to set that up..., :W
I also particularly like the inclusion of the standby power supply.

Now, on the pragmatic side, this isn't a 500W amp. It is apparently running on the same rails as an NC500 modulator or an NC400 amplifier module.

At 8 ohms, I would have rated this as a 200W amp, as north of that the distortion rises rapidly as it enters soft clipping. yeah, it gets up into the 250W range with 0.2% distortion, but that's two orders of magnitude removed from what it normally runs.

At 4 ohms, I would rate this as a 300W/ch amp. Whether there is power supply limiting, or it's just the Class D output stage reaching its limits, I can't say.

The power supply is under designed for high power; you can see this because of the discrepancy of distortion at 2 ohms between 100Hz and 1kHz/6kHz. At the higher frequencies, it can draw the peak power from the caps, and the average/RMS DC power is provided by the supply. At low frequencies, there isn't enough bypass to support LF output, so the SMPS must track the current delivery over the whole waveform. This is a problem with the Pascal modules, too. At 250W at 2 ohms, it is current limiting on low frequency output at 100Hz, likely does the same or worse at lower frequencies. This is what I call an under designed SMPS; B&O ICEPower have similar issues; it's the attempt to balance between cost and performance. In the midrange and high frequencies, the module will eek out 500W at 0.1% THD at 2 ohms. But at low frequencies this is cut in half.

This was a "discussion" (aka argument) that was going on early in 2016 when we were working on a high power solution for the IR Class D stuff; Jun Honda and I both wanted to do it right, with a design that could deliver up to 2.8 kW for two channels at low frequencies, but the marketing pukes wanted to cut the BOM cost in half. In the end, neither solution was done, and fortunately the very ignorant Marketing puke is no longer in the company even, having first transferred to another division, then within a couple of months found a job outside. But the high power project was Lucy'd, as I like to describe it. (Peanuts Charlie Brown and Lucy and the football. ). I'm curious to see if my new assigned project for this year get's Lucy'd or not. I'm pragmatic, as long as the pay checks keep coming, I'll keep working. But as a shareholder and given my global grade, I'll probably have to take some kind of action.

Sorry for the side trip- even though related.

If you consider the 500MP to be a good 250W solution with some headroom, then one is being realistic and won't be disappointed.

[TEK]

Thanks for your input Jon.
So, if considering the NC500MP a 250W@8 ohm amp...
That is not the same as a NC500 + SMPSA700. That would be rated as a 400w@8ohm amp, would it not?

[JonMarsh]

No, it wouldn't. At 8 ohms, the power delivery is limited by voltage- more power requires higher voltage rails. At low impedances, things are limited by current.

In principle, if you have a load invariant amplifier (like a Halcro) if it's 200W at 8 ohms, it's 400W at 4 ohms, and 800W at 2 ohms.

Many/most real world power supplies sag under loading. So low impedance power is not as high, because the rail voltages drop.

The rail voltages set an upper limit on power at 8 ohms. with the MOSFETs that are feasible to use for Class D at this time, (no more than 200V) this limits the rails to about +/- 80V or so, to have some derating room and room for switching spikes. Even the nCore 1200 is only rated for 350W at 8 Ohms. So don't plan on more than 250W from an NC500 at 8 ohms. Of course, 8 ohm speakers are rarely 8 ohms, except some newer Accuton drivers- they're more like 6 ohms. :W