DIYHiFi.org

For the sake of audio
It is currently Sun Dec 17, 2017 2:28 am

All times are UTC




Post new topic Reply to topic  [ 30 posts ]  Go to page Previous  1, 2
Author Message
PostPosted: Thu Dec 04, 2008 6:17 pm 
Offline
Pig
User avatar

Joined: Mon Dec 13, 2004 2:32 pm
Posts: 1325
Don't know about you guys... I wanted to tell this many times but some may find me having an interest in speaking this way, but now I'll risk... So, the whole thread is actually mostly... well, one of those diya style threads.

It is not only about launching an "ultimate DAC", founding it on particular concept of the filter, and not even publishing the filter's frequency response. And not only about leaving the stuff like input stage, output stage, supplies etc, mostly out of the scope of the ultimate project.

The funniest thing about it, and the one I noticed after another guy has asked me how to solve the problem he had with noise, is that the "ultimate TDA1541A DAC" is actually 10 bits DAC, since proposed scheme for external DEM clocking is flawed and just won't work. Six bits simply knocked out. :blink:


Top
 Profile  
Reply with quote  
PostPosted: Thu Dec 04, 2008 7:34 pm 
Offline
Benjamin
User avatar

Joined: Mon Dec 13, 2004 3:29 pm
Posts: 4333
Location: Somewhere other than here.
I don't understand why we are discussing this? I had to go over there to see what is going on.

Shame on all of you! [joke]

Dan:

Are you asking us to 'splain how it works, if we think that it works, or if we think it is a rip-off? Not sure what the point is. Have mercy on a poor old sod and 'splain what the point is.

Jocko


Top
 Profile  
Reply with quote  
PostPosted: Thu Dec 04, 2008 7:55 pm 
Offline
Benjamin
User avatar

Joined: Mon Dec 13, 2004 3:29 pm
Posts: 4333
Location: Somewhere other than here.
Taken directly from Never Connected's site:

"Whilst the supply can be adapted to work with Power amplifier output stages the improvements here were clearly not as great as those achieved on the line level stages with their greater susceptibility to noise and interference."

Oh....is that why.........?

It would be interesting to see exactly how much the DC voltage goes up and down in between charge cycles.

MikeG:

Any experience measuring stuff like that. You've been there, right?

Jocko


Top
 Profile  
Reply with quote  
PostPosted: Thu Dec 04, 2008 8:24 pm 
Offline
Account disabled on owner's request

Joined: Wed Dec 29, 2004 8:31 pm
Posts: 899
This guy uses a cap multiplier, never connected patented a 317 on the output which supposedly was later changed to a discrete regulator in an improved version.

It's a glorified regulator and I think overlooks the practical reality on several facets that adds up to it not being worth a damn in either case.

Seems to me this guy is sidestepping the mention of never connected, alluding to his simpler version, while at the same time trying to claim it as novel. "Yes I "invented" it, unless someone has before", later sidestepping with "mine only uses X parts"...

He's obviously into sales. Maybe contact never connected and see if his version is novel enough for them not to care, could be a laugh.

PS: There's some blurred graphs on the never connected site found in the review section that to my eye shows worse performance with it.


Top
 Profile  
Reply with quote  
PostPosted: Thu Dec 04, 2008 8:37 pm 
Offline
Dog
User avatar

Joined: Tue Dec 14, 2004 12:26 pm
Posts: 4981
Location: Lisbon, Portugal
Jocko Homo wrote:
Taken directly from Never Connected's site:

"Whilst the supply can be adapted to work with Power amplifier output stages the improvements here were clearly not as great as those achieved on the line level stages with their greater susceptibility to noise and interference."

Oh....is that why.........?

It would be interesting to see exactly how much the DC voltage goes up and down in between charge cycles.


Yes, it is really the wrong reason to justify why it won't work with a power amp.
I've heard about this Never Connected PSU some years ago, I think they started using it on phono stages.
Not sure, but whatever.
Batteries do it for me.

_________________
Carlos Filipe

"Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius - and a lot of courage - to move in the opposite direction." Albert Einstein


Top
 Profile  
Reply with quote  
PostPosted: Thu Dec 04, 2008 9:10 pm 
Offline
Sheep

Joined: Thu Jan 13, 2005 10:34 am
Posts: 92
Location: Bath, UK
Been cogitating some more about the 'NC' approach..:

Surely the limit case comes when you increase the refresh frequency to reduce ripple; it may or may not remain a multiple of line frequency.

Now find you have re-invented the SMPS.

Ho hum.


Top
 Profile  
Reply with quote  
PostPosted: Thu Dec 04, 2008 9:38 pm 
Offline
Benjamin
User avatar

Joined: Mon Dec 13, 2004 3:29 pm
Posts: 4333
Location: Somewhere other than here.
Now you have a choice!

High refresh rate = HF noise, just like a switcher. Only with a bigger mains cap.

Low refresh rate = poor 1/f performance.

Carlos:

The reason why my regulator is "elegant" in your words is because of its lack of DC "wobblies". You need a long record time for that to show up on an FFT, but it shows up like poor 1/f. Which is not what you want for a clock.

Now.......if Trichord puts another regulator after it, it probably is ok.

Jocko


Top
 Profile  
Reply with quote  
PostPosted: Thu Dec 04, 2008 9:42 pm 
Offline
Sheep

Joined: Mon Oct 08, 2007 1:37 pm
Posts: 88
Location: Ontario, Canada
Jocko Homo wrote:
I don't understand why we are discussing this? I had to go over there to see what is going on.

Shame on all of you! [joke]

Dan:

Are you asking us to 'splain how it works, if we think that it works, or if we think it is a rip-off? Not sure what the point is. Have mercy on a poor old sod and 'splain what the point is.

Jocko


Bear with me Jocko as I have previously posted that my electronics knowledge is somewhat limited. My only point of interest is if this 'charge transfer' power supply is truly a new idea and an improvement over existing designs. So far responses have claimed that it is a copy of the "Never Connected" design or a possible variation of the flying capacitor psu. The circuit designer has posted enough details so that hopefully those of you well versed in PSU design can give me a specific answer to a specific question(s).

Is this legitimately a new design or a copy of, or rehash of an existing design?

If this circuit works as the circuit designer claims would it be an improvement over existing designs?


Regards,
Dan :banghead:


Top
 Profile  
Reply with quote  
PostPosted: Thu Dec 04, 2008 11:29 pm 
Offline
Benjamin
User avatar

Joined: Mon Dec 13, 2004 3:29 pm
Posts: 4333
Location: Somewhere other than here.
A variation of an existing idea.

Wasn't shaming you............just the rest of the crowd. That's what the [joke] notation is for.

Anyway, now I understand your question. As Martin pointed out, in some ways it is only a SMPS working at 1:1 voltage (not buck or boost) and at a line refresh rate.

While it may keep AC noise off of the load, there will have to be some switching transients. May be minor, but they will be there. You could say that he is swapping one form of noise for another.

Does that help any?

Jocko


Top
 Profile  
Reply with quote  
PostPosted: Thu Dec 04, 2008 11:57 pm 
Offline
Sheep

Joined: Mon Oct 08, 2007 1:37 pm
Posts: 88
Location: Ontario, Canada
Jocko Homo wrote:
A variation of an existing idea.

Wasn't shaming you............just the rest of the crowd. That's what the [joke] notation is for.

Anyway, now I understand your question. As Martin pointed out, in some ways it is only a SMPS working at 1:1 voltage (not buck or boost) and at a line refresh rate.

While it may keep AC noise off of the load, there will have to be some switching transients. May be minor, but they will be there. You could say that he is swapping one form of noise for another.

Does that help any?

Jocko


Yes. Because of my modest background in electronics I'm easily drawn into the claim of a 'new idea'. I depend on the more knowledgeable members of this forum to expose those that are claiming to have 'reinvented the wheel'!

Regards,
Dan :grin:


Top
 Profile  
Reply with quote  
PostPosted: Fri Dec 05, 2008 12:10 am 
Offline
Benjamin
User avatar

Joined: Mon Dec 13, 2004 3:29 pm
Posts: 4333
Location: Somewhere other than here.
OK. Now that we/I know you a bit better, we/I can structure our response more in line with your needs.

Putting all of that aside, you see a lot of nonsense being peddled on forums as "new", "revolutionary", or pick any other ridiculous adjective.

Jocko


Top
 Profile  
Reply with quote  
PostPosted: Fri Dec 05, 2008 1:13 am 
Offline
Account disabled on owner's request

Joined: Wed Dec 29, 2004 8:31 pm
Posts: 899
Because it switches isn't enough to consider this a smps.. this is something much dumber having some of a smps con's against it and really, not a single potential advantage.


Other reasons it sucks is because guess what, it _is_ connected. Noise can and will bleed its way through any which way it likes and there's nothing here preventing that from happening at all.

Mains can vary by a regulated 20%, before ever looking at sudden transients than can be pretty wild. All this will still "imprint" itself during the "charging" phase, and since cap multipliers dont' regulate but merely smooth/filter, and likely better at HF ripple rejection than anything mains related, it will do a fine job of passing these v variations through.

Of course the switching is a potential source of creating even more noise than it pretends to remove.

Never connected is a more refined and still more complex version of the same that still falls prey to a lot of the above to some extent or other, so there's really no justifying it at all wrt a well implemented standard supply. As such in the end it amounts to selling you an overly complex bog standard "regulator"/"filter" that could be far better implemented if it was only something that didn't pretend to cure issues while actually failing to address them.

So, it is a joke in either case, like selling you a car without any tires and claiming it will never get any wear, and whether it's new or not is really only relevant to their respective lawyers.

I'm not sure but this could be the first time I've ever seen a pure marketing gimmick nicked so cleanly.


Top
 Profile  
Reply with quote  
PostPosted: Fri Dec 05, 2008 1:33 pm 
Offline
God

Joined: Thu Dec 30, 2004 10:50 pm
Posts: 998
Location: Minnesota
I haven't been to the NC sight, but, the circuit looks like a simplified early version of a PFC controller. Which would be cool, if there was something in the circuit that actually controlled the pfc.

I have done a lot of measuring this type of stuff. At the lab. Wish I had the gear... Got an extra $50K-$100K anyone?

A line measurement is done a couple different ways; the DUT is plugged into a box, this box injects noise into the mains to the DUT and you monitor the DUT for bad things. Then you classify what "class" your product is in. This ranges anywhere from a complete reboot to nothing happens. Doesn't mean anything to us.

For a fancy measurement the same as above is done except a spectrum analyzer is connected to the output. Jocko, you could do this with the gear you have. You would just have to inject a signal at the input. I'm saving my pennies to get an analyzer for this very purpose.

A burst test; the AC cord of the DUT is inserted into a long tube. This tube couples a signal via EMI to the cord. caveat, RFI and capacitive can also be done. Then monitor the output of the DUT.

For immunity testing for EN/FCC the goal is to not have the DUT fail in some way. Actually measuring the output is not common to obtain a certification. But, when designing, we do measure to get a base line of what's really happening.

As pointed out the problem with these NC methods is that noise is still injected. You can't get away from all the other coupling methods in the real world. Any C in the switch (as mentioned), coupling through close proximity wires, blah, blah. Swapping problem A for problem B.

There is also nothing shown about earth. Any idea how much noise comes in via the earth line? More than one would think. This is assuming that earth is connected like it should be.

I've always wanted to try a pfc chip at the front of a supply. Someday.

The best way to get a handle on this, for minimal cost, is to check out the standards for medical grade gear and hospital AC power requirements. Somewhere I've got the standards. I think. If I can find them... Hospitals can not have more than 20 mohm R between any earth connection and any other earth connection. This alone is huge. Think it through. It does no good to have all the fancy stuff for filtering then have an earth line that's injecting mV of noise right into your box. Actually measured this kind of level. Or, to have an earth floating volts above where it should be. Measured this too.

I prefer still, until I can prove/design otherwise, to use the methods presently available; a balanced transformer which does help drop common mode noise, the standard AC filtering methods, etc.

I recently had to solve a problem like this at work. A capacitive load cell connected to a meter. The meter runs on 24VDC. The load cell is inside a chassis with a 3 phase brushless motor. The servo drive running the motor is connected to 480V. Ouch. Full scale readout is 20K pounds, the reading would vary 3-5 pounds. Not good enough for the ME designing this product.

I used one of my regulators (not plugging anything) to power the meter, then added snubbers at the input filter and output until the meter reading was stable. The same methods known for a long time for filtering/decoupling/bypassing.

Mike


Top
 Profile  
Reply with quote  
PostPosted: Sat Dec 06, 2008 11:31 pm 
Offline
Sheep
User avatar

Joined: Sat Jul 16, 2005 8:39 pm
Posts: 56
Location: Moonee Ponds Victoria
For those interested the Never Connected patent is
US 7092268

Regards
James

_________________
Image


Top
 Profile  
Reply with quote  
PostPosted: Mon Sep 29, 2014 3:16 pm 
Offline
Pig

Joined: Wed Jun 29, 2005 6:34 pm
Posts: 1068
Location: East Coast USA
I guess some revelations take a while.

http://www.6moons.com/industryfeatures/vinnierossi/1.html

Not saying it doesn't or can't work - I'm open minded - but it did seem familiar.

I wonder what the patent office will think.


Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 30 posts ]  Go to page Previous  1, 2

All times are UTC


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group