«

I have two good old harddrives. I cannot place them in my Compaq
Presario.
I need a good quality USB/UDA. I ahve one but is is a reral piece of
junk. Ayone know where I could get a quality USB/UDA?

Thanks

They are all pretty much the same, and they generally all work ok. You
may have a bad one ... the "build quality" of these is poor. Out of
about 5, I've had one with a bad USB/IDE converter (which I fixed; the
USB/IDE chip was poorly soldered) and one with a bad power supply (it
superficially worked, but failures occurred whenever I used it).

Best place to buy them is on E-Bay unless you need one fast, because
most of them ship from China or Hong Kong by boat (2-3 weeks).


teak wrote:
> I have two good old harddrives. I cannot place them in my Compaq
> Presario.
> I need a good quality USB/UDA. I ahve one but is is a reral piece of
> junk. Ayone know where I could get a quality USB/UDA?
>
> Thanks

I agree.

I work at a shop where we use those.
We always buy them 2 at a time, and
when one fails buy 2 more. That's
the way it is with dirt cheap Chinese
electronics. I also agree e-Bay is
the source for the cheapest direct-from-
China products.
--

Barry Watzman wrote:
> They are all pretty much the same, and they generally all work ok. You
> may have a bad one ... the "build quality" of these is poor. Out of
> about 5, I've had one with a bad USB/IDE converter (which I fixed; the
> USB/IDE chip was poorly soldered) and one with a bad power supply (it
> superficially worked, but failures occurred whenever I used it).
>
> Best place to buy them is on E-Bay unless you need one fast, because
> most of them ship from China or Hong Kong by boat (2-3 weeks).
>
>
> teak wrote:
>> I have two good old harddrives. I cannot place them in my Compaq
>> Presario.
>> I need a good quality USB/UDA. I ahve one but is is a reral piece of
>> junk. Ayone know where I could get a quality USB/UDA?
>>
>> Thanks

Attn: Barry Watzman

I generally agree with your posts.
You have a site or any online technical
tutorials a man could take the time to read?

I started in on the forums in '96 with Usenet.
I carry amateur and commercial radio licenses
as well as the A+. My background is obviously
radio, and not as obviously component-level
repair of amateur-grade radio equipment and
consumer-grade electronics in general.

I have hosted online tutorials, but they are
quite specialized and probably of no interest
to you. My old specialty and how I made a
living was CB radio. The mysterious illegal
side of CB radio was where the money was/is,
so that's where I hung my hat. In time I
got tired of it and wanted to go legit.

So here I am--or here I was--I've been legit
for a couple of years now.

I suppose my point is: I'm still interested
in high power. With PCs that's the power
supply. I've been looking for the ideal
tutorial on PC power supply design and testing.
I'm coming up with a mixed bag. Namely that
there are a number of topologies and I (a guy
who is better with linear supplies than switching
supplies) should be able to recognize and be
comfortable with all of them.

Is that true? If so, okay I have some catching
up to do. Maybe you can help me with that as
well. I teach interns that come from a local
vocational school. They all say their time with
me teaches them more (in two months) than the
school did (in six months). I expose them to
the innards of CRT monitors and power supplies.
I want them to be competent and unafraid.

I can keep going, but I'll let it go here.
If you think you can help me in any way, let me
know.

I work for a refurbisher/recycler.

http://www.computersforeveryone.biz

We refurbish donated computer equipment and
sell it cheap or give it away.

We also maintain a free municipal area wireless
network. http://www.wifi101.net
--

================================================== =======
I agree.

I work at a shop where we use those.
We always buy them 2 at a time, and
when one fails buy 2 more. That's
the way it is with dirt cheap Chinese
electronics. I also agree e-Bay is
the source for the cheapest direct-from-
China products.
--

Barry Watzman wrote:
> They are all pretty much the same, and they generally all work ok. You
> may have a bad one ... the "build quality" of these is poor. Out of
> about 5, I've had one with a bad USB/IDE converter (which I fixed; the
> USB/IDE chip was poorly soldered) and one with a bad power supply (it
> superficially worked, but failures occurred whenever I used it).
>
> Best place to buy them is on E-Bay unless you need one fast, because
> most of them ship from China or Hong Kong by boat (2-3 weeks).
>
>
> teak wrote:
>> I have two good old harddrives. I cannot place them in my Compaq
>> Presario.
>> I need a good quality USB/UDA. I ahve one but is is a reral piece of
>> junk. Ayone know where I could get a quality USB/UDA?
>>
>> Thanks

In a nutshell, all switching power supplies work pretty much the same:
The incoming line is full wave rectified and crudely filtered
(capacitor) to DC. The DC is "chopped" on and off through the primary
of a pulse transformer. The secondary (secondaries, in many cases)
provide the output. Since the switching frequency is 20khz to 100khz,
relatively very tiny capacitors provide very high quality pure DC
output. ONE of the outputs is monitored by a circuit that compares it's
output voltage to the desired voltage, and this is used to change the
on/off duty cycle of the primary switching circuitry to keep that
output's voltage within 5% or so of what it should be. The voltages of
the other outputs, if any, are simply a function of the relative turns
ratios of their secondary windings on the pulse transformer (e.g. they
are not individually regulated, per se).

If you ever worked on vacuum tube car radios in the 1950's, you will
recognize this a the 70 year old "vibrator" power supply, only with
solid state electronics replacing the "vibrator", incorporating feedback
and pulse width modulation to [precisely] control the output voltage.

Michael A. Terrell wrote:
> Bill Eitner wrote:
>> I suppose my point is: I'm still interested
>> in high power. With PCs that's the power
>> supply. I've been looking for the ideal
>> tutorial on PC power supply design and testing.
>> I'm coming up with a mixed bag. Namely that
>> there are a number of topologies and I (a guy
>> who is better with linear supplies than switching
>> supplies) should be able to recognize and be
>> comfortable with all of them.
>>
>> Is that true? If so, okay I have some catching
>> up to do. Maybe you can help me with that as
>> well. I teach interns that come from a local
>> vocational school. They all say their time with
>> me teaches them more (in two months) than the
>> school did (in six months). I expose them to
>> the innards of CRT monitors and power supplies.
>> I want them to be competent and unafraid.
>
>
> news:sci.electronics.design is a good place to ask about the various
> types of switching power supplies. There are some schematics of ATX
> type power supplies online where people have reverse engineered them.
>
> <http://www.google.com/search?q=atx+power+supply+schematic+diagram&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-USfficial&client=firefox-a>
>
> The heart of a PC power supply is the controller IC. The data sheets
> are availible online, and have a sample or test circuit to show how it
> works.
>
> The biggest failure mode is defective electrolytics (High ESR) or
> shorted semiconductors.

That's for the input.

I understand the fundamentals. Over the years I've purchased
a few books, but their theory and designs are not specific
to my current interest. I'm looking for PC power supply
specifics so I can put together a troubleshooting and repair
flowchart that I can teach to the interns. Right now we
repair about 50% as bad caps and semiconductors that fail
as shorts are visual. Change out the bad looking parts and
the supply will generally work again. Now I'm looking to
be able to repair 40-45% of the remaining 50% as well as
being able to show the differences between units that are
more likely to be reliable at a given output. Right now
I simply pass along that a difference in weight can matter.
For example, compare a Dell 250 watt unit that weighs twice
as much as a cheap 400 watt (so says the sticker) unit.
Now I want to be able to point out the specific differences.
As time permits I've been working my way through this site:
http://www.smps.us/computer-power-supply.html
So far it looks pretty good.

The other project is the same sort of deal but with LCD
monitors. Right now if it's not a signal processing issue
and online research of the specific brand and model points
to soldering, bad caps, semiconductors in the power supply,
we more often than not repair it. However, that still leaves
a large percentage unrepaired that could likely be repaired
if my understanding was greater.

Thanks again,

Bill Eitner
http://www.computersforeveryone.biz
--

Michael A. Terrell wrote:

>>> The heart of a PC power supply is the controller IC. The data sheets
>>> are availible online, and have a sample or test circuit to show how it
>>> works.
>>>
>>> The biggest failure mode is defective electrolytics (High ESR) or
>>> shorted semiconductors.
>> That's for the input.
>>
>> I understand the fundamentals. Over the years I've purchased
>> a few books, but their theory and designs are not specific
>> to my current interest. I'm looking for PC power supply
>> specifics so I can put together a troubleshooting and repair
>> flowchart that I can teach to the interns. Right now we
>> repair about 50% as bad caps and semiconductors that fail
>> as shorts are visual. Change out the bad looking parts and
>> the supply will generally work again. Now I'm looking to
>> be able to repair 40-45% of the remaining 50% as well as
>> being able to show the differences between units that are
>> more likely to be reliable at a given output. Right now
>> I simply pass along that a difference in weight can matter.
>> For example, compare a Dell 250 watt unit that weighs twice
>> as much as a cheap 400 watt (so says the sticker) unit.
>
>
> That 250 Watt supply may use a lower switching frequency, which
> requires more iron in the transformers. The switching semiconductors
> can be bipolar or Field effect. Bipolar has a higher voltage drop at
> saturation, which requires more heat sinking so weight isn't a great way
> to judge. That 450 Watt supply may use active rectification instead of
> Schottky diodes to reduce waste heat.

There you have it. Or the claimed 400 watt supplies we
bought a case of for no more than $15 a piece are exactly
what we paid for.

I'm at a wall. I need the specific knowledge and tools
to break through. I want to be able to tell the difference
and show my work.

> Compare the original 63 watt IBM PC supply (Made by Astek) to that
> 250 watt Compaq. It is heavier, runs at a much lower frequency and is
> all Bipolar semiconductors. They used an AC fan for cooling to get
> enough air to keep them from dying. Some of the fans were diecast
> aluminum.

And it was only a 63 watt supply?! That can't be right.
The ICs in an IBM PC were lined up row after row.
You're telling me the whole mess ran off of a 63 watt
(and why not 65 or 80 or 100 watt) supply? Maybe 163--
something like that. I always thought the original
supplies were over a hundred watts--because they needed
to be to feed the acre of chips--and switching type
because IBM engineers of the time could make it happen
with improved overall efficiency.

> Some of the other failures are open resistors in the control loop, or
> on the IC used as a window comparator that monitors the individual
> outputs. Shorted ceramic capacitors around the ICs will cause out of
> spec regulation, no output or shorted switching transistors.

Thanks for that. Shorted ceramic capacitors; that's one
I certainly wouldn't look for early on. As an RF repair
tech, ceramic cap failures are the last thing I look for.
I've seen them blown open (cracked open ceramic causing
a larger gap between the plates and reduced capacitance).
Having a ceramic cap fail so as the plates come together
to form a short would be a new one for me.

To me, open resistor means burnt open which we can deal
with now (thankfully).

>> Now I want to be able to point out the specific differences.
>> As time permits I've been working my way through this site:
>> http://www.smps.us/computer-power-supply.html
>> So far it looks pretty good.
>
>
> The data sheets I mentioned will show you how that IC works. How the
> output voltage is fed back to the regulator, and what the internal
> reference voltage is. I have repaired switching supplies for over 23
> years, and had no schematic for 99% of them.

Where did you start/what did you start with?

> Once you understand how
> the circuits work, they are a lot easier to troubleshoot. The only way
> a flowchart will help is if it's for that one model power supply.

Okay. So I need to be teaching circuit understanding.
Not really practical within a 2 month internship but
understandable in this context.

>> The other project is the same sort of deal but with LCD
>> monitors. Right now if it's not a signal processing issue
>> and online research of the specific brand and model points
>> to soldering, bad caps, semiconductors in the power supply,
>> we more often than not repair it. However, that still leaves
>> a large percentage unrepaired that could likely be repaired
>> if my understanding was greater.
>
>
> You can always ask for help on news:sci.electronics.repair

You know what I'll find--the LCD monitor equivalent of
you. And he'll say a generalized flow chart isn't possible
even when I know he could create it (just like you could
with PC power supplies).

The bottom line is: if you can fix a particular device without
a schematic 99% of the time, you can produce a flow chart of
how you do it.

For a long time my gig was CB radio. I was in it for decades
and got so good at some of the common yet popular tasks that
I wrote concrete tutorials that took away all the mystery.
I shared them with a popular site. They are dated now, but
are still there as proof:
http://www.cbtricks.com/members/kd6tas/index.htm

Help me move on.
--

The original IBM-PC supply was made by Zenith (the old TV maker, also
the owner of Heathkit and Zenith Data Systems), not Astek (which did,
however, become a supplier later).

[I worked for Zenith at the time]