CHANGSTAR: Audiophile Headphone Reviews and Early 90s Style BBS

  • December 31, 2015, 11:07:28 AM
  • Welcome, Guest
Please login or register.

Login with username, password and session length
Advanced search  

Author Topic: ASCC where numbers hit the wires  (Read 779 times)

0 Members and 1 Guest are viewing this topic.

johnjen

  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +10/-0
  • Offline Offline
  • Posts: 148
  • Sonicus Blasticus
ASCC where numbers hit the wires
« on: January 23, 2015, 10:52:10 AM »

So is there any interest in discussing what
Available Short-Circuit Current is?


I'm performing a series of tests that measure ASCC and other related ac power delivery metrics and will be relating these to SQ changes.


JJ
Logged
PWD-II w/bridge (lightly modded) - Schiit Ragnarok - The Rok (lightly modded) - HD-800 SAA modded, balanced & hardwired
Shunyata Python & α-Digital & Anaconda XLR - AudioQuest Coffee - Wyrd Schiit - 'Other' cables & Tweaks aplenty
Schiit Mojo (retired for now)

Solderdude

  • Grab the dScope Kowalski!
  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +206/-4
  • Offline Offline
  • Posts: 907
  • No can do skipper, the dScope was terminated
    • DIY-Audio-Heaven
Re: ASCC where numbers hit the wires
« Reply #1 on: January 23, 2015, 11:23:23 AM »

Will you be measuring it on various outlets, extension cords, power cords ?
determining the ASC is essential for calculating wire diameters, lenghts and fuses/circuit breaker properties when designing electrical wiring.

For stereo equipment this would relate to the inrush currents mainly ?
The ASC will generally be MUCH higher than that of the power transformer in audio devices.

Interested to see if there is a relation and how that relation can be shown.

Will you be measuring the actual currents (using an oscilloscope to evaluate actual wave forms and peak currents) in the mains cables of individual equipment in 'operational' state as well ?
Logged
Use your ears to enjoy music, not as an analyser.

johnjen

  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +10/-0
  • Offline Offline
  • Posts: 148
  • Sonicus Blasticus
Re: ASCC where numbers hit the wires
« Reply #2 on: January 23, 2015, 11:57:01 AM »

Will you be measuring it on various outlets, extension cords, power cords ?
determining the ASC is essential for calculating wire diameters, lenghts and fuses/circuit breaker properties when designing electrical wiring.

For stereo equipment this would relate to the inrush currents mainly ?
The ASC will generally be MUCH higher than that of the power transformer in audio devices.

Interested to see if there is a relation and how that relation can be shown.

Will you be measuring the actual currents (using an oscilloscope to evaluate actual wave forms and peak currents) in the mains cables of individual equipment in 'operational' state as well ?

I have measured it on my 'standard' house power distribution system as well as my dedicated runs for the audio and computer system using 9 different cables.

That's the interesting thing about ASCC it essentially measures short duration, high amperage availability of the branch circuit.

Alas I wish I had a 'calibrated' and tweako O-scope setup, but all I have are 2 handheld AC circuit load testers.  One is the Extech CT70 and the other is a General CA10.  Neither is calibrateable nor do their readings always agree on these peak current readings.

But they both show the same trends and show a relationship between the increases in current dumping ability and wire gage and type of circuit being tested.  Which tells me that while I can't 'trust' the specific numbers I can use the trends and patterns as confirmation of the effects, especially when 2 separate units confirm these patterns and trends.

As for measuring all of this in realtime on my gear that is in my system, again no, I'm not set up to be able to 'tear into' my Rok nor my PWD and monitor these actual values in situ.

But the results are interesting none the less.
At least I think so.

JJ
Logged
PWD-II w/bridge (lightly modded) - Schiit Ragnarok - The Rok (lightly modded) - HD-800 SAA modded, balanced & hardwired
Shunyata Python & α-Digital & Anaconda XLR - AudioQuest Coffee - Wyrd Schiit - 'Other' cables & Tweaks aplenty
Schiit Mojo (retired for now)

johnjen

  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +10/-0
  • Offline Offline
  • Posts: 148
  • Sonicus Blasticus
Re: ASCC where numbers hit the wires
« Reply #3 on: January 25, 2015, 04:11:30 AM »

Hmmmmm…

So let me try this approach…

If you do the math, one cycle of a 60Hz power feed takes ≈ 16.67 ms.  Which means that half a cycle takes ≈ 8.33 ms. 
This 8ms is significant for a number of reasons and amongst them is the I^2t measurement of fuse behavior. 
It also plays a role in the way that the diode switched power supplies in our gear ‘operates’.

Lemme explain.
Fuses are rated by how much current they can pass before giving their all and opening up.  One such test (I^2t - ampere squared seconds) measures the maximum amount of current that the fuse will pass in 8ms before dying.   
Which means they are rating each fuse for how much current it will pass during 1/2 of a cycle of our 60Hz ac power.

To wit…
“A pulse of current is applied to the fuse, and a time measurement is taken for melting to occur. If melting does not occur within a short duration of about 8 milliseconds (0.008 seconds) or less, the level of pulse current is increased. This test procedure is repeated until melting of the fuse element is confined to within about 8 milliseconds. The purpose of this procedure is to assure that the heat created has insufficient time to thermally conduct away from the fuse element. That is, all of the heat energy (I^2t) is used, to cause melting. Once the measurements of current (I) and time (t) are determined, it is a simple matter to calculate melting I^2t. When the melting phase reaches completion, an electrical arc occurs immediately prior to the “opening” of the fuse element.

Clearing I^2t = Melting I^2t + arcing I^2t

The nominal I2t values given in this publication pertain to the melting phase portion of the “clearing” or “opening”.  Alternatively the time can be measured at 10 times of the rated current and the I2t value is calculated like above.”

From © 2014 Littelfuse • Fuseology Selection Guide  www.littelfuse.com

Additionally the diode bridge in our power supplies only conduct (passes power to the rest of the power supply) when the secondary voltage from the power transformer is greater than the existing voltage of the downstream power supply itself.  Which means that the diodes switch ‘on’ for only a portion of that 8ms.  And for this portion of this thought experiment we’ll use a 50% duty cycle and say that the diodes are ‘on’ for ≈ 4ms and then ‘off’ for ≈ 4ms, then back ‘on’ for ≈ 4ms etc.
Which means that the incoming ac power is only actually powering the device 1/2 of the available time. 
And it’s doing so in ≈ 4ms bursts followed by ≈ 4ms of being switched off, ad infinitum.
Which means that the amount of current the device requires is ‘pulsed’ thru the transformer and back out the ac supply on that particular branch circuit.

This is what ASCC is actually measuring. 
It ‘simulates’ a dead short between hot and neutral and/or hot and neutral and ground.  And gives us a reading of the maximum amount of current for that 8ms ‘pulse’, if it even lasts that long (depending on which type of circuit protection is involved).

ASCC is a means of determining how much short duration current the particular branch circuit being tested, can actually deliver, before the sparks start flying.

And it all happens in 8ms, or less.
Well, my guess is, our amps and dacs etc. are probably pulling current in ≈ 2-3ms bursts with ≈ 5-6ms ‘off’ timing. 
Mostly because I doubt that the power supply is ‘operating’ at roughly 1/2 the transformers secondary voltage. 
It’s probably closer to 70-80% (or more) before the diodes switch on, which shortens the ‘on’ cycle time, and lengthens the ‘off’ cycle times.
This reduction of ‘on’ time means that the magnitude of the pulse of current must increase, given the same demand for power to operate the device.
It also means that the power supply is now pulling all it's needed current during only ≈ 20-30% of the available time, instead of ≈ 50%.

Fortunately many fuses are able to dump increasing amounts of current as the duration of the pulse shortens, especially as the pulse duration extends down into the 4-1ms range.

This is why ASCC has caught my attention, why I’m investigating further, and why the preliminary results I’ve gathered thus far support this perspective. 
It also portends what we can do to improve the SQ of our systems. 
If so motivated.

JJ
« Last Edit: January 25, 2015, 04:37:55 AM by johnjen »
Logged
PWD-II w/bridge (lightly modded) - Schiit Ragnarok - The Rok (lightly modded) - HD-800 SAA modded, balanced & hardwired
Shunyata Python & α-Digital & Anaconda XLR - AudioQuest Coffee - Wyrd Schiit - 'Other' cables & Tweaks aplenty
Schiit Mojo (retired for now)

Solderdude

  • Grab the dScope Kowalski!
  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +206/-4
  • Offline Offline
  • Posts: 907
  • No can do skipper, the dScope was terminated
    • DIY-Audio-Heaven
Re: ASCC where numbers hit the wires
« Reply #4 on: January 25, 2015, 07:57:29 AM »

I have already done some real life measurements a while ago.
Not regarding fuses but currents in rectifiers, so on the secondary side.
It is good to know that the currents drawn on the primary side are the same factor lower than the voltage factor a transformer has.

This may be an interesting read.
https://diyaudioheaven.wordpress.com/tutorials/power-supplies/
scroll down to the rectifier section (5) and don, t forget to read the snubber section (6) as well.
Forget about my objectivistic vieuws and read over that in the knowing I don, t have any audio knowledge at all.

The measurements are made with single phase rectifiers, with bridge rectifiers the , on time,  and currents will be shorter.
Logged
Use your ears to enjoy music, not as an analyser.

johnjen

  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +10/-0
  • Offline Offline
  • Posts: 148
  • Sonicus Blasticus
Re: ASCC where numbers hit the wires
« Reply #5 on: January 25, 2015, 10:09:45 AM »

That's a wonderful read…
Thanks for that.  :thumb !

I just skimed thru most of it, and I will take more time and digest it more throughly.

My only question thus far is how would you scale those current measurements up to those in actual use in say a Mojo amp.  I use that (and please feel free to substitute…) amp because it has a 'strong' class A section and it's all solid state so it operates more 'efficiently' than tubes…

And
"It is good to know that the currents drawn on the primary side are the same factor lower than the voltage factor a transformer has."
Could you say this another way?

Thanks again for that link. 
Moar food for thought!  :thumb


JJ
Logged
PWD-II w/bridge (lightly modded) - Schiit Ragnarok - The Rok (lightly modded) - HD-800 SAA modded, balanced & hardwired
Shunyata Python & α-Digital & Anaconda XLR - AudioQuest Coffee - Wyrd Schiit - 'Other' cables & Tweaks aplenty
Schiit Mojo (retired for now)

Solderdude

  • Grab the dScope Kowalski!
  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +206/-4
  • Offline Offline
  • Posts: 907
  • No can do skipper, the dScope was terminated
    • DIY-Audio-Heaven
Re: ASCC where numbers hit the wires
« Reply #6 on: January 25, 2015, 11:30:42 AM »

Edit: I added examples of a 160VA trafo with an 80W load with single- and dual-wave rectifiers in the article: https://diyaudioheaven.wordpress.com/tutorials/power-supplies/.
Numbers below have been altered accordingly to those measurements.


Of course things change when a bridge rectifier is used. They change for the better as the capacitor is charged double the amount of time.
For 60Hz the currents will probabaly even be a factor 0.9 smaller as the capacitor is charged is a faster rate namely every 8.3ms instead of 10ms.

To estimate the peak currents that are present with the rag will thus depend on power usage and rating of the transformers used.
The power transformer is 400VA and the smaller one 75VA so lets assume 500VA in trafo and at max ouptut power 500W is drawn.
This will never be continuous but peak only when listening to music at almost distortion levels the actual power drawn may be around 300W or lower.
Acc to Schiit's power ratings the currents drawn are 4.5Arms (at peak loads) and 0.7Arms in idle (while driving HP's the output load won't increase much, say 1.2Arms at the most).

I assume dual phase rectifiers are used so the peak currents will be lower and the pulse duration will be shorter compared to the example.
In the 160VA example a 1:3.5 ratio was found for 50Hz in between peak currents (12.5A) and drawn continuous current (3.55A).
At 60Hz the time between 'replenishments' is 20% shorter so the ripple will be smaller and the currents will be smaller.

Assuming worst case (inrush and max power drawn) of 500W and 115V mains voltage this would be 4.4A rms. Because the peak currents are a factor 3.5 higher in my tests (50Hz) the factor will likely be slightly above 3.2 for 60Hz. It would be reasonable to assume a 400VA trafo will act quite similar to a 160VA trafo. The peak currents will thus be around 14A peak when 500W is drawn.
When driving HP's at deafening levels around 3.5A peak.

Don't have a Ragnarok so cannot check it but figure these numbers won't be that far off.

When you have a poor mains wiring (say 1 Ohm) the 14A peak would drop the input voltage at the trafo by 14V peak while the mains 115V has peak voltages of 162V.
The rms voltage on the mains input will thus 'clip' close to '106V RMS' and will be causing harmonics on the mains due to flattening of the sine wave.

For us Europeans (50Hz, 230V) the current will be about twice as low and because our peak voltage is higher as well (325V) the penalty is less percentage wise as the drop would only be about 8V.
Of course in the USA the wiring in the house is thicker and fuses are rated higher.
In the EU most fuses are 16A, some 10A in normal houses.



« Last Edit: January 26, 2015, 01:43:40 PM by Solderdude »
Logged
Use your ears to enjoy music, not as an analyser.

johnjen

  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +10/-0
  • Offline Offline
  • Posts: 148
  • Sonicus Blasticus
Re: ASCC where numbers hit the wires
« Reply #7 on: January 26, 2015, 10:34:02 AM »

Excellent!
Thanks for that!

Concise and dense…
Just like I like…   :)p7

I need to cogitate it for a bit, but I have a couple of questions burbling up from the depths…   


JJ :thumb
Logged
PWD-II w/bridge (lightly modded) - Schiit Ragnarok - The Rok (lightly modded) - HD-800 SAA modded, balanced & hardwired
Shunyata Python & α-Digital & Anaconda XLR - AudioQuest Coffee - Wyrd Schiit - 'Other' cables & Tweaks aplenty
Schiit Mojo (retired for now)