SC DSP Musicolour - Page 2

Do you have a question? Post it now! No Registration Necessary

Translate This Thread From English to

Threaded View
USB supply scan on ABSE

"Andy Wood"
Quoted text here. Click to load it


** LOL  !!

I have put a scan of the schematic on ABSE for you.


....  Phil



Re: USB supply scan on ABSE
Quoted text here. Click to load it
The transistor to hold off the FET seems a about to me.
Rheilly P

Re: USB supply scan on ABSE

Quoted text here. Click to load it

For those without easy usenet binary access:
http://www.alternatezone.com/stuff/USBsupply.jpg

Come on, step right up, spot the mistake in the circuit!

Dave.



Re: USB supply scan on ABSE

Quoted text here. Click to load it

can I have another go Dave?
Please can I, can I? Pretty please :-)

Strangely, it is obvious when you stop looking too hard.

Don...



--
Don McKenzie

Site Map:            http://www.dontronics.com/sitemap
We've slightly trimmed the long signature. Click to see the full one.
Re: USB supply scan on ABSE
Quoted text here. Click to load it

In fact, there is a wee bit of a mess, and I think it all came about
because of what I said in my first guess.

You need to read the text of the article to knit it together.

Don...


--
Don McKenzie

Site Map:            http://www.dontronics.com/sitemap
We've slightly trimmed the long signature. Click to see the full one.
Re: USB supply scan on ABSE

Quoted text here. Click to load it
access:
http://www.alternatezone.com/stuff/USBsupply.jpg
Quoted text here. Click to load it


6V in - .7V across the diode = not enough headroom for the regulator?

Cheers,

Al






Re: USB supply scan on ABSE

Quoted text here. Click to load it

Oops, it's a Schottky. So @ 1A, the LDO needs 5.5V, but the specified
diode will typically drop .6V - which means out of 6V, the regulator
gets 5.4V (slightly less, given the FET drop). Still not the greatest
design for 1A. How much current is this thing supposed to supply?

I get the feeling I'm missing something obvious.

Cheers,

Al



Re: USB supply scan on ABSE
Quoted text here. Click to load it

I would suggest there is nothing wrong with the mechanics Al, even
though it may appear marginal. I went down the same path initially.

I doubt that the drive would draw an amp.

But the original design problem was excessing heat, so changes were made.

A clue:
I'm putting my money on the result of those changes not being completed
correctly. :-)

Don...


--
Don McKenzie

Site Map:            http://www.dontronics.com/sitemap
We've slightly trimmed the long signature. Click to see the full one.
Re: USB supply scan on ABSE

Quoted text here. Click to load it

OK, please keep in mind I have only seen the scan, so I have no idea
what this thing is supposed to do. But here's my guess:

It used to use a higher voltage power supply and a 7805. The '05 got
too hot, so they changed to an LDO and a 5V supply. But they forgot to
read the datasheet correctly. It states

"The minimum output capacitance required to maintain stability is 22
B5%F (this value may be increased without limit)"

And we have a .1uF cap on the output, so the whole thing may
oscillate.

Do I win a prize?

Cheers,

Al

Re: USB supply scan on ABSE
Quoted text here. Click to load it

Does Al win a prize Dave?

Don...




--
Don McKenzie

Site Map:            http://www.dontronics.com/sitemap
We've slightly trimmed the long signature. Click to see the full one.
Re: USB supply scan on ABSE

Quoted text here. Click to load it

Al does indeed win the prize!
But I suspect Don knew all along...

Classic mistake with LDO's, and this one in particular is notorious for it.
Yes, 22uF minium is required, so there is a very high likelihood this sucker
is going to oscillate.
So not only did the previous design have an error (overheating), the new one
has a nasty one too. All on the simplist of circuits, unbelievable. The text
even says the 0.1uF is there to ensure stability!

Dave.



Re: USB supply scan on ABSE



Quoted text here. Click to load it

It's a classic error wihen people drop in  LDO regulators because it's a slowish
pnp collector output instead of an npn emitter output. For similar reasons note
that 79xx's also require more local decoupling than 78xx's.

Graham


Re: USB supply scan on ABSE



Quoted text here. Click to load it

I didn't get a prize on a.b.s.e for saying that !        :-(

Graham


Re: USB supply scan on ABSE
On Sat, 31 May 2008 03:08:08 -0700 (PDT), Al  Borowski

Quoted text here. Click to load it

Spot on! The original design required 9V AC/DC and used a 7805.

USB Power Injector (Issue 193, Published: 18 October, 2004):
http://www.siliconchip.com.au/cms/A_102685/article.html
http://us1.webpublications.com.au/static/images/articles/i1026/102685_5mg.jpg

Quoted text here. Click to load it

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

Re: USB supply scan on ABSE



Quoted text here. Click to load it
access:
http://www.alternatezone.com/stuff/USBsupply.jpg
Quoted text here. Click to load it

It's a schottky. 1N5817 would be better though.

Graham


Re: USB supply scan on ABSE

Quoted text here. Click to load it

I don't know about you guys, but I haven't seen too many PC motherboards with
USB type B outputs.

Type A, yes, but not Type B.

dmm


Re: USB supply scan on ABSE

Quoted text here. Click to load it
USB type B outputs.

Then again, a type A to type B cable should fix that.

Quoted text here. Click to load it


Re: USB supply scan on ABSE

Quoted text here. Click to load it
USB type B outputs.
Quoted text here. Click to load it

yes, no problems there, the cable design I would consider a standard
intercept.

Don...


--
Don McKenzie

Site Map:            http://www.dontronics.com/sitemap
We've slightly trimmed the long signature. Click to see the full one.
Re: USB supply scan on ABSE
dmmilne_REMOVE_@ozemail.com.au says...
Quoted text here. Click to load it
USB type B outputs.
Quoted text here. Click to load it
Huh?

I don't see a problem.
The input side is type B, just like any peripheral, the output is type
A, just like the PC.

In other words, the input looks just like a peripheral, the output just
like the PC.

Where's the problem?

Ray

Re: SC DSP Musicolour
Quoted text here. Click to load it


    I just got my notification that my SC subscription's expired ... I
kinda doubt that I'll be renewing it. Nothing it it has grabbed my
attention (in a good way) for a very long time.
    This thread confirms what I've been thinking. :-(



Bob


Site Timeline