Balanced Power Supply

What people are working on at the moment
Post Reply
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8863
Joined: Mon May 21, 2007 9:42 pm

#76

Post by Paul Barker »

Yes but rather than wind some of each winding on each leg wind the primary and secondary far apart so that there is only coupling through the core.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
Nick
Site Admin
Posts: 15706
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#77

Post by Nick »

Or use an EI or C with a split bobbin.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8863
Joined: Mon May 21, 2007 9:42 pm

#78

Post by Paul Barker »

Nick wrote:Or use an EI or C with a split bobbin.
good idea.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8863
Joined: Mon May 21, 2007 9:42 pm

#79

Post by Paul Barker »

Pulled out of another thread

Quote Moi

"Off to a slight tangent, the most likely direction my efforts to generate an independent clean supply would take should I ever get it off the ground, may well be a DC motor speed controlled (via stabilised DC power supply absolutely loaded up with solid state) driving a 3 phase generator, (full rotary balanced 3 phase generation, not just pulling a third phase with a capacitor and then passing the output through a motor to try and equalise it, which the usual rotary converters do) and make my B+'s 3 phase choke input. Heater supplies would have to be from another supply as I wouldn't want to pull an imbalance from the phases."

Back on terra firma. What actually needs clean mains?

Just wandering if the valve amp B+ say with valve rectification and regulation perhaps some of each type Series/shunt. How much overlaid mains distortion gets through if any?

Just a thought.

Can see it upseting a dht heater supply and an AK43XX DAC for sure.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
figlet108
User
Posts: 66
Joined: Mon Jun 17, 2013 9:23 am
Location: Macclesfield

#80

Post by figlet108 »

Last night I set up a test environment with my BPS powering ALL my kit and with the options of 2 AG1500 regenerators before and/or after the BPS.

Voltage of BPS under full load:
In----Out
230 - 245 (AG1500 inputting 230v to BPS reported 50% load; AG1500 on output being fed 245v by BPS reported 40% load)
245 - 260 (BPS fed directly by mains supply)

The voltage of the idle BPS when fed 230v was 250, so it's only dropped 5v when under load.

I'll call Airlink and just see what their view is.
Jason
User avatar
jack
Thermionic Monk Status
Posts: 5493
Joined: Wed Dec 29, 2010 8:58 pm
Location: ɐılɐɹʇsnɐ oʇ ƃuıʌoɯ ƃuıɹǝpısuoɔ
Contact:

#81

Post by jack »

Please remember thar most small UPS systems are of the "off-line" type, whereby they are simply connected to the mains directly unless the mains falls out of pre-set parameters, in which case it switches to generated AC from the battery. As soon as the mains is OK again, it switches back - the battery is kept continually charged from the mains until its needed.

An off-line UPS is therefore only driven from batteries when the mains fails.

The AC generated is synthesised and often harmonic-rich, i.e. not very pure.

More expensive UPS systems are "on-line", whereby the mains is rectified and used to charge a bank of SLAa, the DC output of which are connected to the inverter which is connected to the UPS output.

The only time an on-line UPS connects its output directly to the mains is when either the UPS fails on the unit is deliberately switched into bypass for maintenance (e.g. replacing the SLAs).

"Better" UPSs are on-line and detail the harmonic content of their output - really clever ones, like large systems used for data centres, can balance 400V 3 phase systems on the input side whilst catering for unbalance 3 phase loading on the output side. Such systems are pricey, but very impressive (each 19 inch rack can weigh in at over 1 tonne).

So, have a look at what your UPS is actually doing first :)
Vivitur ingenio, caetera mortis erunt
User avatar
figlet108
User
Posts: 66
Joined: Mon Jun 17, 2013 9:23 am
Location: Macclesfield

#82

Post by figlet108 »

Hi Nick, thanks for the info about UPSes, very interesting.

However, the AG1500 is a different type of product - have a read

It regenerates the mains 100% of the time and was designed more for A/V applications than as a standard UPS for the IT industry. Think of it as a poor man's PS Audio P10 or one of the PurePower products.
Jason
User avatar
jack
Thermionic Monk Status
Posts: 5493
Joined: Wed Dec 29, 2010 8:58 pm
Location: ɐılɐɹʇsnɐ oʇ ƃuıʌoɯ ƃuıɹǝpısuoɔ
Contact:

#83

Post by jack »

figlet108 wrote:Hi Nick, thanks for the info about UPSes, very interesting.

However, the AG1500 is a different type of product - have a read

It regenerates the mains 100% of the time and was designed more for A/V applications than as a standard UPS for the IT industry. Think of it as a poor man's PS Audio P10 or one of the PurePower products.
Its not new in any way by their own admission. Its a standard on-line UPS - they say as much.

They use phrases like "They are built upon tried and tested online double conversion technology" (see above) and "effects caused by electrostriction" (think "humming output transformer").

No decent on-line UPS uses an input or output transformer - its relatively simple to do the whole thing in solid-state.

Their USP (not "UPS"!) seems to be using a PWM fan. Yipee. Name me a PC or UPS that doesn't have that nowadays...

Also, they charge 500 quid for this thing - an excellent APC unit (BR1500GI) costs 200 quid less - http://www.apc.com/products/resource/in ... _watts=200
Last edited by jack on Tue Sep 17, 2013 4:50 pm, edited 1 time in total.
Vivitur ingenio, caetera mortis erunt
User avatar
figlet108
User
Posts: 66
Joined: Mon Jun 17, 2013 9:23 am
Location: Macclesfield

#84

Post by figlet108 »

Ok Nick. It's a bog standard on-line UPS.
It's still feeding the BPS 230v of regenerated AC.
Jason
User avatar
figlet108
User
Posts: 66
Joined: Mon Jun 17, 2013 9:23 am
Location: Macclesfield

#85

Post by figlet108 »

Spoke to Airlink today and they confirmed that the BPS is faulty.
They said that when fed 230v at idle it should output 236v and under full load should hit 230v. So it's going back.
However, I must say that their after sales service is second to none - really excellent company to deal with.

Despite the high output voltage I'm currently running AG1500->BPS->AG1500 and it does sound particularly sweet tonight.
I'm not going to bother spending time doing comparisons until the BPS gets replaced though...
Jason
User avatar
Ali Tait
Eternally single
Posts: 4373
Joined: Fri Jun 08, 2007 8:10 pm
Location: Galashiels

#86

Post by Ali Tait »

Good news Jason.
User avatar
figlet108
User
Posts: 66
Joined: Mon Jun 17, 2013 9:23 am
Location: Macclesfield

#87

Post by figlet108 »

Ali Tait wrote:Good news Jason.
Yes, from what I'm hearing tonight I'm really looking forward to getting the BPS back and doing some comparisons with different configurations.
Although I have to say, I've only just got the system back up after many many weeks while redecorating the living room and I've missed proper music so much that everything sounds sublime.
I wasted yesterday evening listening to out of phase 'noise' unable to figure out why it sounded shit :)

Anyway, regarding the very interesting discussion above above El transformers vs toroidal, I came across this by Pedja Rogic
You guys probably already know all this but an eye opener for me...
Jason
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8863
Joined: Mon May 21, 2007 9:42 pm

#88

Post by Paul Barker »

Intriguing that it took him a load of costly and timely research to discover what I pointed out above as the bleeedin obvious. But I suppose if your job is to sit at a desk and design something it better seem you did something with your time.

I don't quite appreicate why the lowe powered transformers have poorer bandwidth and the difference on the graph doesn't appear significant. The fact that the roll off is in keeping with the our needs is good enough. there would be other advantages of the higher powered transformer to our power supplies as Nick G already showed. We need a low source impedance.

I can testify that power torroids make excellent parafeed output transformers for mid frequency and HF. There is insufficient inductance for LF performance. But for anyone bi amping they are a credible Push pull or parafeed option for vertually no money. The graph demonstrated this excellently.

However the EI graph also demonstrated that for table top radio requirements an EI filament supply transformer can also be used as an output transformer. Many people including myself used a filament transformer for our first ever sound making circuit we hooked up on the bench, and we were wowed over to valvae amplifiers forever after. My first experience was with the ECC82 into the EL34 triode connected with a filament transformer on the output. It sounded far better than anything solid state I have ever heard, to my ears.

I'll stick with EI for balanced power, if I eventually decide it is doing something of greater value than a low pass filter which can be done cheaper in other ways.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
jack
Thermionic Monk Status
Posts: 5493
Joined: Wed Dec 29, 2010 8:58 pm
Location: ɐılɐɹʇsnɐ oʇ ƃuıʌoɯ ƃuıɹǝpısuoɔ
Contact:

#89

Post by jack »

What I didn't understand was the obsession with having the mains tx doing all the suppression.

Sensible people, IMHO, use an standard Schaffner EMC/EMI (or similar) filter and therefore don't have to give a monkey's about the frequency response of the main tx (other than hat its good for 50 or 60Hz...)

Maybe I'm missing something :)
Vivitur ingenio, caetera mortis erunt
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8863
Joined: Mon May 21, 2007 9:42 pm

#90

Post by Paul Barker »

Interesting view Nick, but then do you really need the transformer at all? If you are going to have one, why not make it optimal?

Regarding the filters, use one, you hear an improvement, cascade two or three you hear an improvement at every step. So make your transformer also a low pass filter, you also hear an incremental improvement.

I can assure you a single EMI commercial filter sounds pants compared to a cascade of them in our world.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
Post Reply