advertisement


vertically mounted NCC200s?

hammeredklavier said:
Can you explain a bit further? I'm not sure what you mean. As before, I'm not saying you're wrong, just trying to understand the problems!
Click to expand...

Arkless covers this below very nicely.

Arkless Electronics said:
Hint to understanding the issue: If a wire has 0.1 Ohm resistance and 10A AC is flowing then you have 1V AC generated between the ends of this wire... "Ground" is now 1V AC....
Click to expand...

Plus you have front end charging currents that create small voltages in the 0V return of the front end which gets amplified by the gain of the amp. This will make it sound like a mess as it is possible that the speaker currents (across the speaker 0V return resistance) are phase shifted in relation to the front end. Both will create separate voltages across their respective 0V return resistances, some phase shifted in relation to each other and thus the front end and output notional 0V are different, and not only vary in relation to each other, but are to some degree frequency dependent due to speaker reactance.
 
a.palfreyman said:
Arkless covers this below very nicely.



Plus you have front end charging currents that create small voltages in the 0V return of the front end which gets amplified by the gain of the amp. This will make it sound like a mess as it is possible that the speaker currents (across the speaker 0V return resistance) are phase shifted in relation to the front end. Both will create separate voltages across their respective 0V return resistances, some phase shifted in relation to each other and thus the front end and output notional 0V are different, and not only vary in relation to each other, but are to some degree frequency dependent due to speaker reactance.
Click to expand...

You are nearly ready to leave the temple Grasshopper:)

When there are also several rail decoupling caps.. and maybe one or two regulated supplies for front end and other stages.. with their own decoupling caps... it gets like 3D chess! :eek::mad::confused::) It fecks my head right over I can tell you! I'm never certain that any such more complicated grounding layout is truly optimum.... It can only be different blends of compromises:(

Here's a true story of interesting grounding... On the MF A1 and A100 there are two 6" ish lengths of green and yellow striped 0.75mm CSA ish wire in the phono stage area which are looped and hot melt glued to the PCB. They go to ground, as did the original PCB tracks which are cut. There was a bad error in the ground layout of the PCB which caused hum on phono. These act as humbuckers and give a fraction of a mV of opposite phase hum to cancel the ground hum from the original layout! Love the lateral thinking on that one.. even if it was to cure the original fault. Don't know if it was TdP (probably was) or someone at MF who came up with that. It was fait accompli when i joined MF.
 
Arkless Electronics said:
When there are also several rail decoupling caps.. and maybe one or two regulated supplies for front end and other stages.. with their own decoupling caps... it gets like 3D chess! :eek::mad::confused::) It fecks my head right over I can tell you! I'm never certain that any such more complicated grounding layout is truly optimum.... It can only be different blends of compromises:(
Click to expand...
:D
 
Christ. I'm surprised the amps didn't come with a note specifying that power rails must be 150.46mm of 0.75mm2 wire, 31.37mm from the chassis at all times, 20.8oC and 1atm pressure! :D Five cable ties or six?! :eek:
 
Ah, that's where you went wrong then. You need SEVEN cable ties....;)
All joking apart, keep earth runs short as possible and bundle speaker return in with +/- PSU wires to keep radiated fields low. Attach all earth wires to a single point between the last caps in the bank, or to the centre leg of a 'T' (if you have just one PSU for both amps as you will have at least 4 returns) with the top of the tee being the link between the PSU caps.
 
a.palfreyman said:
Ah, that's where you went wrong then. You need SEVEN cable ties....;)
All joking apart, keep earth runs short as possible and bundle speaker return in with +/- PSU wires to keep radiated fields low. Attach all earth wires to a single point between the last caps in the bank, or to the centre leg of a 'T' (if you have just one PSU for both amps as you will have at least 4 returns) with the top of the tee being the link between the PSU caps.
Click to expand...
Thanks buddy. As you say, jesting aside, I'm just astonished not to have read a lot more discussion of wiring layout in the past. If it's THAT critical you'd think there'd be a fairly well established optimum layout for these Naim-type amps, with the wire lengths and gauges also specified. If the impedances really are that critical then surely you need to account for that when roughing out your build, even in typical one-box applications.

I'm most interested to know how Naim made the two-box layout work for them in the NAP500. I don't think I've ever even seen one, but if memory serves that has the transformer, rectifier and smoothing capacitors in a separate box and the amplifiers in another. Connected with a lead that looks like it'd feed a 500kW induction motor! :D
 
Never seen one either. FAR to rich for me. I'd think that they will have local decoupling tight to each of the amp boards with both 0V returns to that local decoupling per channel, all fed by a fat lead.
 
As I mentioned earlier, Naim make the 2 box approach work by putting the regulators very close to the power amp. The power star can then be moved from the main smoothing caps to the power star of the + and - regs. Ideally the speaker return should go to the point of lowest impedance.
 
hammeredklavier said:
Thanks buddy. As you say, jesting aside, I'm just astonished not to have read a lot more discussion of wiring layout in the past. If it's THAT critical you'd think there'd be a fairly well established optimum layout for these Naim-type amps, with the wire lengths and gauges also specified. If the impedances really are that critical then surely you need to account for that when roughing out your build, even in typical one-box applications.
Click to expand...

Some good reading here:
https://hifisonix.com/ground-loops/
 
It's not an easy picture to get! I've been designing, modding and building amps for more than 30 years and agree with Jez - it's a brain fryer.

FWIW, I don't entirely agree with the hifisonix info in terms of where to return the speaker 0V to. IMO he ignores "the point of lowest impedance" and instead returns to the local decoupling on the pcb. Since everyone on DIYaudio follows this method, I'm obviously wrong ;).
Try both and see which you prefer the sound of.

I am a fan of the approach Naim take in the NAP 200. If you study the pcb you will see a signal star (both channels) and a power star per channel, the latter two connected to the former by a long thin track! I ussed this approach in my NCC300 and it sounded much better than dual mono with the speaker return to the local pcb decoupling.

See the last pic here:
https://www.aliexpress.com/item/1005002646628929.html
Hint: what's the circular blob of copper connected to the GND wire?
 
I haven't had a Naim amp for a long time now, but as I recall, they were quite hissy and even hummy (by the standards of mainstream audio equipment these days, which is on the whole amazingly quiet).

Many thanks for the hifisonix link btw, I've only had time for a very quick flash through it but it seems very illuminating! :)
 
hammeredklavier said:
I haven't had a Naim amp for a long time now, but as I recall, they were quite hissy and even hummy (by the standards of mainstream audio equipment these days, which is on the whole amazingly quiet).

Many thanks for the hifisonix link btw, I've only had time for a very quick flash through it but it seems very illuminating! :)
Click to expand...

I never said Naim sounds good! ;)
However, the grounding arrangement of the NAP200 is well thought out IMO. The clone I built sounded decent and had zero audible hum or noise. I think Naim preamps is where most of the noise comes from... and there are good reasons for that.
 
Only front and rear panels are aluminium according to the link and, like anything from RS, it's way over priced.

Modushop/HiFi2000 have the market to themselves if you don't wish to buy Chinese. You can specify ali or steel top and bottom panels for many of their cases. They will also sell these separately and I had to buy in replacement ali panels for a steel case of theirs recently after a change of tack on a job meant I needed to use the case as a heatsink after all...
 
Arkless Electronics said:
Only front and rear panels are aluminium according to the link and, like anything from RS, it's way over priced.

Modushop/HiFi2000 have the market to themselves if you don't wish to buy Chinese. You can specify ali or steel top and bottom panels for many of their cases. They will also sell these separately and I had to buy in replacement ali panels for a steel case of theirs recently after a change of tack on a job meant I needed to use the case as a heatsink after all...
Click to expand...
Hmm. It actually doesn't say what the top and bottom panels are made of does it.

Nothing's easy...! :-D
 
hammeredklavier said:
Hmm. It actually doesn't say what the top and bottom panels are made of does it.
Click to expand...

I was somewhat opposed to the ModuShop offerings as I felt they are somewhat overpriced for what they are. Having just bought three with a "summer" discount added to the benefits of buying three cases I think the price paid qualifies as "fair".

The Dissipante have heatsinks which are clearly more than adequate for Avondale products.

Regards

Richard
 
Richard Lines said:
I was somewhat opposed to the ModuShop offerings as I felt they are somewhat overpriced for what they are. Having just bought three with a "summer" discount added to the benefits of buying three cases I think the price paid qualifies as "fair".

The Dissipante have heatsinks which are clearly more than adequate for Avondale products.

Regards

Richard
Click to expand...
Yeah I'll probably just order one. It's probably overkill for the NCC200s, but then, the whole project is an exercise in overkill so what the hell! :D
 
Richard Lines said:
I was somewhat opposed to the ModuShop offerings as I felt they are somewhat overpriced for what they are. Having just bought three with a "summer" discount added to the benefits of buying three cases I think the price paid qualifies as "fair".

The Dissipante have heatsinks which are clearly more than adequate for Avondale products.

Regards

Richard
Click to expand...

I thought the same and checked out the cost of making my own... the cost of the aluminium is truly horrific! They will be buying it in much bigger quantities and getting considerable discount of course but to buy in enough sheet aluminium to make say 4 cases would have cost about 60% of the cost of the finished cases from Modushop!!:eek::eek:

Add the cost of getting the metal cut and drilled etc to professional standard by a local engineering firm and I don't believe you could make them for any less.... then there's the considerable labour and hassle involved!

Also in the case (sic) of the larger Dissipante cases with huge heatsinks then similar heatsinks bought in from RS etc are not at all competitive on price compared with the Modushop offerings.

TLDR; I also thought they were charging a fortune for "empty boxes" but after this exercise I have to regard the prices as very fair.
 
You must log in or register to reply here.


advertisement


Back
Top