advertisement


Mains conditioning

Arkless Electronics said:
"Let's hear your expert opinion then... I'm sure you wouldn't express an opinion unless you have the ability to design power supplies yourself and know all about it now would you?"

andyr said:
I would use a linear PS - not a SMPS - for an analogue component (tube or ss). SMPSes go better with digital components.

Andy
Click to expand...

Aha... about as I thought.. In other words you know about as much about electronics as I do about embroidery.... Thanks for your sage advice.
Click to expand...

My 'sage advice', O snarky one, is based on the experiments I have done ... and what I have observed across several people's systems. :p

A couple of years ago, I introduced a new, very sophistocated TT motor controller into my system. This needed a 48v DC supply - so was supplied with a Meanwell SMPS. My (biased) PoV then, was that SMPSes were bad and I should be able to deliver a better result with a linear PS. So I organised some listening tests with some friends - including the designer of the speed controller, Steve Tuckett ("Number9"). We listened to a couple of tracks using 5 different PS arrangements for the motor controller. The results were as follows - arranged in order of our evaluation of which one we thought delivered better sound (5 is best sound):
1. Meanwell SMPS
2. linear PS (designed by someone who designs amps for a living)
3. 48v SLA supply - 2 rows of 4x 12v SLAs, in parallel to provide a low Zout
4. linear PS with an isolating transformer plus hash filter between it and the wall socket
5. Meanwell SMPS with an isolating transformer plus hash filter between it and the wall socket.

We had assumed that the linear PS would deliver the best sound and were all surprised that the best sound (for this digital component) came from the Meanwell SMPS - but only after we had "cleaned it up" by putting an isotran plus hash filter between it and the wall socket. Steve explained this by saying the speed of a SMPS synergises better with a digital component than a (slow!) linear PS does.

For any analogue component, however, my view is that mfrs who use SMPSes do so for two reasons - neither of them because it sounds better!:
1. cost savings, and
2. virtue signalling (look at us ... we're green, saving energy by using an efficient SMPS!).

For analogue components, better sound will come from linear PSes, IMO. And it would seem other people think the same way - for instance the Hypex 800w plate amps on my subs use a conventional linear PS - not a SMPS.

Andy
 
Mike Reed said:
Are you hard-wired too?
Click to expand...

No, not hard wired, but most of the system is powered by a Hydra style mains distribution block.

PIaTpaYl.jpg
 
Arkless Electronics said:
"Let's hear your expert opinion then... I'm sure you wouldn't express an opinion unless you have the ability to design power supplies yourself and know all about it now would you?"
Click to expand...


andyr said:
A couple of years ago, I introduced a new, very sophistocated TT motor controller into my system. This needed a 48v DC supply - so was supplied with a Meanwell SMPS. My (biased) PoV then, was that SMPSes were bad and I should be able to deliver a better result with a linear PS. So I organised some listening tests with some friends - including the designer of the speed controller, Steve Tuckett ("Number9"). We listened to a couple of tracks using 5 different PS arrangements for the motor controller. The results were as follows - arranged in order of our evaluation of which one we thought delivered better sound (5 is best sound):
1. Meanwell SMPS
2. linear PS (designed by someone who designs amps for a living)
3. 48v SLA supply - 2 rows of 4x 12v SLAs, in parallel to provide a low Zout
4. linear PS with an isolating transformer plus hash filter between it and the wall socket
5. Meanwell SMPS with an isolating transformer plus hash filter between it and the wall socket.
Andy
Click to expand...

Arkless Electronics said:
Aha... about as I thought.. In other words you know about as much about electronics as I do about embroidery.... Thanks for your sage advice.
Click to expand...
Bugger me Arkless, never knew you were so enthusiastic about a bit of Double Flemish Cross Stitch. Do you favour silk or wool? My missus does a bit, you should come over and have a chat with her about it.
 
stevec67 said:
Bugger me Arkless, never knew you were so enthusiastic about a bit of Double Flemish Cross Stitch. Do you favour silk or wool? My missus does a bit, you should come over and have a chat with her about it.
Click to expand...

"Double Flemish Cross Stitch" with silk ... sounds positively pornographic! :D

Andy
 
andyr said:
My 'sage advice', O snarky one, is based on the experiments I have done ... and what I have observed across several people's systems. :p

A couple of years ago, I introduced a new, very sophistocated TT motor controller into my system. This needed a 48v DC supply - so was supplied with a Meanwell SMPS. My (biased) PoV then, was that SMPSes were bad and I should be able to deliver a better result with a linear PS. So I organised some listening tests with some friends - including the designer of the speed controller, Steve Tuckett ("Number9"). We listened to a couple of tracks using 5 different PS arrangements for the motor controller. The results were as follows - arranged in order of our evaluation of which one we thought delivered better sound (5 is best sound):
1. Meanwell SMPS
2. linear PS (designed by someone who designs amps for a living)
3. 48v SLA supply - 2 rows of 4x 12v SLAs, in parallel to provide a low Zout
4. linear PS with an isolating transformer plus hash filter between it and the wall socket
5. Meanwell SMPS with an isolating transformer plus hash filter between it and the wall socket.

We had assumed that the linear PS would deliver the best sound and were all surprised that the best sound (for this digital component) came from the Meanwell SMPS - but only after we had "cleaned it up" by putting an isotran plus hash filter between it and the wall socket. Steve explained this by saying the speed of a SMPS synergises better with a digital component than a (slow!) linear PS does.

For any analogue component, however, my view is that mfrs who use SMPSes do so for two reasons - neither of them because it sounds better!:
1. cost savings, and
2. virtue signalling (look at us ... we're green, saving energy by using an efficient SMPS!).

For analogue components, better sound will come from linear PSes, IMO. And it would seem other people think the same way - for instance the Hypex 800w plate amps on my subs use a conventional linear PS - not a SMPS.

Andy
Click to expand...

Such complete tosh that I can't be bothered to go through it point by point but can't resist the biggest howler as it's just so funny to anyone with technical knowledge... "Steve explained this by saying the speed of a SMPS synergises better with a digital component than a (slow!) linear PS does" comedy gold:D
 
YNWOAN said:
No, not hard wired, but most of the system is powered by a Hydra style mains distribution block.
Click to expand...

Thanks, Mark. Still hard wired, but I thought maybe into radials as I have. Looks like the feed (supply) cable is the same as the 6 load cables.
 
No, there are IEC plugs at the equipment ends and a 13A plug to the mains. The six output cables are all Belden 19364 and the input cable is a short length of Belden 83803. The 83803 is physically thinner but actually has thicker conductors thanks to the thinner Teflon insulation. It looks similar in the photos because it has some black braid over it to hide it’s rather garish red colour.
 
Years ago, my TV will make a lot of picture noise whenever my wife had the hair dryer on. The picture noise disappeared after installing the Isotek mains conditioner. This lead me to believe if I can see the difference when using the mains conditioner then it must be able to make a difference to my audio system as well. I'm happy to say it does, sounds are tighter and background noise have been reduced.
 
Did the Hi-Fi affect the hair dryer? Seriously though, it is more likely that the shielding on leads can cause problems, not being good enough!
 
Good question. I've installed a new electric consumer unit since then. I think the old unit had a separate fuse for the upstairs and downstairs ring. The hair dryer was used upstairs and the HiFI and TV on the downstairs ring.
 
s33k3r100m said:
Good question. I've installed a new electric consumer unit since then. I think the old unit had a separate fuse for the upstairs and downstairs ring. The hair dryer was used upstairs and the HiFI and TV on the downstairs ring.
Click to expand...
It probably was on another ring, this is normal. The old CU's with fuses rather than RCD's used to typically use 1 30A ring circuit upstairs, one for the electric cooker and socket, maybe one for the rest of the kitchen, and one for the rest of the downstairs. A shower got its own circuit.

It's easier to separate the upstairs and downstairs, for obvious reasons. 30A is ample for all the upstairs rooms (other than a shower) even in a big house because it's enough for 7.5kW and electrical items in bedrooms are generally low power. You can run a couple of 2kW fan heaters and 2 hairdryers on 7.5kW, and you are going some to be doing that continuously.
 
The old fuse box (surely they started to be called consumer units when MCBs and RCDs were fitted ?) would have had separate fuses, though my memory, which may well be suspect, was that 15 amp fuse wire was usually used, at least for upstairs, and 5 amp for lighting *. Regardless, I asked because the hair-dryer interference may not have been mains borne. However, old installations with fuses were a bit dicky compared with c.u.s. Shower and cooker circuits were usually radials, not rings.

* Or maybe I just used that wire to renew blown fuses.
 
5A was the lighting loop for sure, it allowed them to run lighter (1.0mm?) cable, so it was cheaper. Still is of course. I have only ever seen 30A rings (for sockets), these are wired 2.5mm T+E. If you only have a radial circuit (not a ring) then 1 x 2.5 T+E gives you 20A safe load, which may explain 15A fusing on the circuit. In older circuits 15A upstairs may well have been adequate, back then there were fewer appliances in use.

I wonder what they will do with the advent of cheap LED lighting? It would make sense to run low voltage cable, say about 12V @ 5A, round the lighting circuits when LED becomes the default. Safer, less fire risk, no shock hazard, much cheaper and easier to install, a definite win-win. I wouldn't be surprised to see it come along in the next few years.
 
SELV is already used for things like wet zones in bathrooms under IEE 17th ed (my shower extractor fan is 24VAC with remote isolating transformer, for example) but I think lighting circuits are likely to simply stay full mains V AC for a long while yet; since the reg requirements for mains powered circuits aren;t likely to actually save on the wiring cost, protection + cabling location+grade requirements (it's still mains powered ultimately, expect rccd protection as a minimum, no real change); and the cost in the wiring is the copper and placing it right, not the insulant (would barely change at all for install rated stuff)

The real 'why' is more simply - because a lot of mains LED lights actually use series strings of many, many tiny COB LEDs to actually generate the output - so typically still need >100vdc + min to light at all. The growth in LED lighting efficacy has gone the 'Moores law' approach - multiplicity of tiny emitters and smaller lighter cheaper leds fabbed in gangs (that look like one old style smt led) put in series to natively run at HV - over brute force, 'just one good low-voltage LED 'big emitter' ... it is actually more efficient.
 
...and many of the more advanced high power multiple LED's are powered by SMPS designed for the specific task... for what it's worth.. You're never more than a few yards from an SMPS these days!
 
You must log in or register to reply here.


advertisement


Back
Top