audio recording on IC -help wanted

[John Jardine]

Sean Mathias <seanm@prosolve.com> wrote in message
news:idjmhvcmsbvon6tb9a47o1e6bu03mmusl4@4ax.com...

I have a number of power transformers I have picked up and am looking
to use one for a specific application. Most are completely unmarked.
It is generally easy enough to determine the pinnings through
continuity and resistance checks, likewise by connecting it to a
source, the voltages are easily determined. What I am struggling with
is how to determine the current it is designed or rated for, any
suggestions?

Thanks,

Sean Mathias

Usually reckon 25Watts per pound weight, (open frame or toroid).

 

[Robert Monsen]

Don't electrolytic caps require a bias voltage to keep them healthy? Seems
like wiring two back to back would cause problems with this.

Regards,
Bob Monsen

"DarkMatter" <DarkMatter@thebarattheendoftheuniverse.org> wrote in message
news:e6glhv0e1ok6bj9r3m1uk8o0fa8ii2b4h0@4ax.com...

On Sat, 19 Jul 2003 07:01:51 GMT, Dan Fraser <dmfraser@sbcglobal.net
Gave us:

A non polarized capacitor of the same capacitance and voltage rating can
replace a polarized capacitor in almost every application with NO
modifications.

Actually, most NP versions are nearly twice the size physically as
their polarized counterpart, given the same voltage spec. That would
mean that many applications would not fit the bigger cap if the
original was tightly packed in the initial design.

In fact the diode trick won't work because current has to flow both in
and out of a capacitor.

However, why bother as polarized caps are far cheaper than unpolarized.

Two. Back to back. Quick and cheap. NP caps cost less than two,
however, in normal buying circumstances. They are harder to find in
single quantities though, which would make your statement probably
correct. They really ARE two caps back to back inside. That is what
an NP cap is.

 

[Dieter Wiedmann]

"Lewin A.R.W. Edwards" schrieb:

Sorry to ask this type of question again, but I have no luck driving
AMP's search engine, or Digi-Key... I'm looking for the part#(s) of
the 8-pin and 9-pin connectors that go to the old AT-style 17-pin
power connector found on AT-form-factor motherboards. (IOW I want the
connector that is on the power supply, because I'm making a custom
cabling harness to power a single-board computer that has this style
of power input connector).

They are made by Molex. P/N are 15-48-0106 and 90331-0001.


Regards,
Dieter

 

[DarkMatter]

On Mon, 21 Jul 2003 20:27:11 GMT, "Robert Monsen" <nospam@nowhere.com>
Gave us:

Don't electrolytic caps require a bias voltage to keep them healthy? Seems
like wiring two back to back would cause problems with this.

Non-polarized electrolytics ARE physically two caps of the given

value in a back to back configuration, in a single package.

 

[Spehro Pefhany]

On Tue, 22 Jul 2003 03:09:54 GMT, the renowned "Robert Monsen"
<nospam@nowhere.com> wrote:

"DarkMatter" <DarkMatter@thebarattheendoftheuniverse.org> wrote in message
news:ml2phvs30r6o4j035u8mj9r0r337of9uon@4ax.com...
[original comment deleted]
Non-polarized electrolytics ARE physically two caps of the given
value in a back to back configuration, in a single package.

I've also heard something like this (notably in Grob's "Basic Electronics"),
but I still don't believe it is strictly true.

If a non polarized electrolytic capacitor consists of two back-to-back
(series connected) simple electrolytic capacitors, then a reverse charge
will be put across one of the capacitors during any charge cycle. I was
under the distinct impression that subjecting any electrolytic to more than
2 V of reverse bias would result in a dangerous situation, possibly leading
to venting of electrolyte and explosion. Connecting the capacitors in
parallel would likewise result in a reverse bias.

The common gets charged to a negative or positive voltage.

From what I can determine by searching the web (references below), I believe
that nonpolarized electrolytic capacitors are actually electrolytic
capacitors in which the dielectric oxide layer has been formed on both the
anode and cathode during the manufacturing process. In a polarized
electrolytic capacitor, only the anode has this dielectric oxide layer
(actually, the cathode has it also, except its too thin to withstand more
than about 2V. This, I gather, is for protection in the case of an
accidental reverse bias.)

Absolutely correct. They look identical to any other electrolytic
inside. It's all in the processing of the foil to create the oxide
layer before they are assembled. Cut one apart and see (I have).


Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com

 

[Ross Herbert]

You will have to determine your requirements but I believe you will find
what you are looking for here
http://www.ciitech.com/products.qpl.asp?sel_item_1=6

Ross Herbert

jipihorn wrote:

Hello everybody !

I search for 20 mil spec relays like these :
http://jipihorn.free.fr/Projets%20en%20cours/Doc/MilspecRelais.jpg with
exactly the same can, ( but voltage does not need to be the same). I've
bought some of these but the parcel was lost and at the same time I made the
printed boards, so I'm obliged to find the same kind of component...
So, is there some people that know where to find that ? Or somebody who have
this model and can trade with other kind of milspec relays...

Many thanks in advance...

J.P.

 

[Ross Herbert]

Particularly look at M39016/6


Ross Herbert wrote:

You will have to determine your requirements but I believe you will find
what you are looking for here
http://www.ciitech.com/products.qpl.asp?sel_item_1=6

Ross Herbert

jipihorn wrote:

Hello everybody !

I search for 20 mil spec relays like these :
http://jipihorn.free.fr/Projets%20en%20cours/Doc/MilspecRelais.jpg with
exactly the same can, ( but voltage does not need to be the same). I've
bought some of these but the parcel was lost and at the same time I made the
printed boards, so I'm obliged to find the same kind of component...
So, is there some people that know where to find that ? Or somebody who have
this model and can trade with other kind of milspec relays...

Many thanks in advance...

J.P.

 

[Lewin A.R.W. Edwards]

Dieter Wiedmann <Dieter.Wiedmann@t-online.de> wrote in message news:<3F1C70A7.706AEDF3@t-online.de>...

the 8-pin and 9-pin connectors that go to the old AT-style 17-pin

They are made by Molex. P/N are 15-48-0106 and 90331-0001.

Vielen Dank That's what I needed to know.

 

[Gary Tait]

On 22 Jul 2003 07:38:40 -0700, ghazan@ghazan.haider.name (Ghazan
Haider) wrote:

I am designing an embedded system, and although I'm acquainted with
the MCUs RAM and friends, I dont quite know the difference between an
ethernet controller and transciever ICs. Which one is the minimum
required for implementing the communications (I know both are not
required).

The MAC.

The controller is a bus level interface and controller between the CPU
bus and the MAC, if the controller has no built in MAC.

On a related note, how much processing power would an average ethernet
IC require from the CPU/MPU? Is 20MIPS (I know this benchmark is very
crude) 32-bit IC with TCPIP implemented sufficient to get say a cs8900
or rtl8019as working? Are there simpler ethernet ICs available?

Also has anyone implemented TCPIP ipv4 onto FPGAs doing NAT etc
without any CPUs?

 

[Thierry Nouspikel]

"Lewin A.R.W. Edwards" wrote:

I'm encountering problems when writing to the card. Namely, the proper
number of bytes gets written at the right address, but these bytes
appear as 0x00, no matter what I actually wrote!

a) Did you erase the block before writing?

Yes, I did.

b) Are you ABSOLUTELY SURE BEYOND ANY POSSIBLE DOUBT that you are
within timing spec? I had a similar problem when implementing SSFDC
and after sitting down with two oscilloscopes I worked out that I
wasn't holding the write strobe long enough.

If anything, I'm holding it for much longer than needed. The TI-99/4A is
an awfully slow computer. Could being too slow be a problem?

The other non-standard thing I'm doing is that I do not hold down ALE
and CLE in between the various data write/read operations. They are both
low *during* the data operation, as required, but may go high between
two bytes. Shouldn't be a problem since these lines are sampled during
the WE* pulse, when they *are* low...

Thanks for your help, anyhow.

Thierry

 

[Lewin A.R.W. Edwards]

b) Are you ABSOLUTELY SURE BEYOND ANY POSSIBLE DOUBT that you are
within timing spec? I had a similar problem when implementing SSFDC

If anything, I'm holding it for much longer than needed. The TI-99/4A is
an awfully slow computer. Could being too slow be a problem?

I shouldn't think this is a problem.

The other non-standard thing I'm doing is that I do not hold down ALE
and CLE in between the various data write/read operations. They are both

I'm not sure about whether this might be a problem or not, since it's
not characterized.

In case it will be a help, here is the algorithm I use (this is
cut-n-pasted block comments from my working code)

{earlier... in card reset code}
// set WE# high, CE# high, CLE low, ALE low

{page write code}
// send CE# low, CLE high
// put Sequential Data Input command on data bus
// send WE# low
// bring WE# high
// bring CE# high
// bring CLE low
// bring CE# low
// bring ALE high
// put column address (always 0x00) on data bus
// bring WE# low
// bring WE# high
// put low byte of page address on data bus
// bring WE# low
// bring WE# high
// put page address mid byte on data bus (bits 8-15 of address)
// bring WE# low
// bring WE# high

if (cardsize >=64) {
// put page address high byte on data bus (bits 16-24 of address)
// bring WE# low
// bring WE# high
}

// bring ALE low
for (i=0;i<pagesize;i++) {
databus = buffer;
// bring WE# low
// bring WE# high
}

// put Program command on bus
// bring CE# low and CLE high
// bring WE# low
// bring WE# high

// wait for busy pin to go inactive

// put Read Status command on data bus
// bring ALE low, CLE high
// bring WE# low
// bring WE# high
// bring CLE low

// wait for busy pin to go inactive

// bring RE# low
// bring RE# high

// bring CE# high to deactivate card

 

[Johnboy]

Why don't you just pick up a freeby dead supply with
long leads to use as your custom harness.

"Dieter Wiedmann" <Dieter.Wiedmann@t-online.de> wrote in message
news:3F1C70A7.706AEDF3@t-online.de...

"Lewin A.R.W. Edwards" schrieb:

Sorry to ask this type of question again, but I have no luck
driving
AMP's search engine, or Digi-Key... I'm looking for the part#(s)
of
the 8-pin and 9-pin connectors that go to the old AT-style 17-pin
power connector found on AT-form-factor motherboards. (IOW I want
the
connector that is on the power supply, because I'm making a custom
cabling harness to power a single-board computer that has this
style
of power input connector).

They are made by Molex. P/N are 15-48-0106 and 90331-0001.


Regards,
Dieter

 

[Stig Carlsson]

Some details: A normal polarized electrolytic capacitor has two foils,
anodic and cathodic, separated by paper impregnated with electrolyte. The
anodic foil has an oxide layer which is the dielectric of the capacitor. The
thickness of the oxide layer is propotional to the voltage applied when the
foil is manufactured. There is also an oxide layer on the cathodic foil, but
it is very thin, about 1 Volt.

The oxide layer can only support voltage in one direction.

If the voltage is applied correctly it vill be across the anodic foil and
all is well. If applied backwords the anodic foil will be shorted out, a
large current will pass and the cathode foil will start to grow its own,
thicker oxid layer. This will lead to gas and heat being produced, and the
capacitor will explode if the availible current is big enough.

In a nonpolarized capacitor there are two anodic foils so it does not matter
which of the foils gets the plus or minus connection, one one the foils will
always be connected correctly.

Regards
Stig Carlsson

 

[Per-Ake Andersson]

In article <bfk6iv$1ne$1@news.eunet.yu>, "Pesovic Uros" <p_uros@eunet.yu> wrote:

Can anyone tell me value of a peak current for this MOSFET-s?
Continous current is 49A(for IRFN44Z) and 53 A(for IRFN48Z)

Missprint för IRFZ44N and IRFZ48N ???

http://www.irf.com

/Per-Ake

--
Remove "extra" in my e-mail adress !!

 

[Jim Meyer]

"Robert Monsen" <nospam@nowhere.com> wrote in message news:<6W1Ta.115655$ye4.85380@sccrnsc01>...

From what I can determine by searching the web (references below), I believe
that nonpolarized electrolytic capacitors are actually electrolytic
capacitors in which the dielectric oxide layer has been formed on both the
anode and cathode during the manufacturing process. In a polarized
electrolytic capacitor, only the anode has this dielectric oxide layer
(actually, the cathode has it also, except its too thin to withstand more
than about 2V. This, I gather, is for protection in the case of an
accidental reverse bias.)

Thus, I believe its probably not a good idea to put electrolytic capacitors
back to back, as suggested previously.

Regards
Bob Monsen

This comes up about once every two or three years.

It *IS* OK to put two polarized caps back-to-back to make an AC
cap out of them. The reason is that there is a tiny bit more leakage
in the reverse biased cap. That leakage charges the "properly" biased
cap in the right direction to keep it happy when the current reverses.
Since the circuit is symetrical, each cap's leakage protects the
other.

Do a test with real capacitors (or even a SPICE simulation). You
will find that after only a couple of cycles that each cap will have a
DC voltage across it that is of a magnitude and polarity such that it
will be completely protected from reverse voltages.

You can guild the lilly with a couple of diodes. One diode
across each cap in a direction that will prevent the cap's voltage
from going more than a diode's drop reverse biased. The diodes are
not necessary but may give you that belt and suspenders feeling of
confidence.

Jim

 

[Lewin A.R.W. Edwards]

There is only a little PCB inside with a logo on it that looks like a
cloverleaf (though a bit more square). Only two numbers on it (besides

The board is made by Casio, and those marks are alignment marks. I
assume that the connector is broken? It's *probably* a JAE part (Japan
Aviation Electronics).

Look for a cradle on ebay, it's your easiest route to success. Or use
infra-red if possible.

 

[Robert Monsen]

I just tried to do your experiment, since it seemed absurd to me (how could
one side charge without passing charge the wrong way through to the other
side?) 14v of AC (off of a step down transformer) across two back to back
(positive side out) 100uF 25v polarized capacitors.

I should have been wearing safety goggles. I was just about to hook up the
oscilloscope to check out this mysterious DC bias, when one of the 100uf
capacitors exploded right in front of me. I've never seen that before! Its
amazing how much stuff sprayed out all over the place. They were both brand
new. The electrolyte got all over my new tek probes and superstrip!

Regards,
Bob Monsen

"Jim Meyer" <jmeyer@nektonresearch.com> wrote in message
news:21ede509.0307230831.68589965@posting.google.com...

"Robert Monsen" <nospam@nowhere.com> wrote in message
news:<6W1Ta.115655$ye4.85380@sccrnsc01>...
From what I can determine by searching the web (references below), I
believe
that nonpolarized electrolytic capacitors are actually electrolytic
capacitors in which the dielectric oxide layer has been formed on both
the
anode and cathode during the manufacturing process. In a polarized
electrolytic capacitor, only the anode has this dielectric oxide layer
(actually, the cathode has it also, except its too thin to withstand
more
than about 2V. This, I gather, is for protection in the case of an
accidental reverse bias.)

Thus, I believe its probably not a good idea to put electrolytic
capacitors
back to back, as suggested previously.

Regards
Bob Monsen

This comes up about once every two or three years.

It *IS* OK to put two polarized caps back-to-back to make an AC
cap out of them. The reason is that there is a tiny bit more leakage
in the reverse biased cap. That leakage charges the "properly" biased
cap in the right direction to keep it happy when the current reverses.
Since the circuit is symetrical, each cap's leakage protects the
other.

Do a test with real capacitors (or even a SPICE simulation). You
will find that after only a couple of cycles that each cap will have a
DC voltage across it that is of a magnitude and polarity such that it
will be completely protected from reverse voltages.

You can guild the lilly with a couple of diodes. One diode
across each cap in a direction that will prevent the cap's voltage
from going more than a diode's drop reverse biased. The diodes are
not necessary but may give you that belt and suspenders feeling of
confidence.

Jim

 

[Robert Monsen]

Actually, once I started cleaning up the mess from the explosion, I realized
that BOTH had exploded. The reason I didn't notice is that one only sprayed
electrolyte. The CAN exploded on the other one. This is NOT a good way to
hook up capacitors, and will serve as a lesson to me to be careful with
polarized electrolytics during circuit failure modes. It took less than 10
seconds to blow it up.

Regards,
Bob Monsen

"Robert Monsen" <nospam@nowhere.com> wrote in message
news:qPDTa.130389$Ph3.16509@sccrnsc04...

I just tried to do your experiment, since it seemed absurd to me (how
could
one side charge without passing charge the wrong way through to the other
side?) 14v of AC (off of a step down transformer) across two back to back
(positive side out) 100uF 25v polarized capacitors.

I should have been wearing safety goggles. I was just about to hook up the
oscilloscope to check out this mysterious DC bias, when one of the 100uf
capacitors exploded right in front of me. I've never seen that before! Its
amazing how much stuff sprayed out all over the place. They were both
brand
new. The electrolyte got all over my new tek probes and superstrip!

Regards,
Bob Monsen

"Jim Meyer" <jmeyer@nektonresearch.com> wrote in message
news:21ede509.0307230831.68589965@posting.google.com...
"Robert Monsen" <nospam@nowhere.com> wrote in message
news:<6W1Ta.115655$ye4.85380@sccrnsc01>...
From what I can determine by searching the web (references below), I
believe
that nonpolarized electrolytic capacitors are actually electrolytic
capacitors in which the dielectric oxide layer has been formed on both
the
anode and cathode during the manufacturing process. In a polarized
electrolytic capacitor, only the anode has this dielectric oxide layer
(actually, the cathode has it also, except its too thin to withstand
more
than about 2V. This, I gather, is for protection in the case of an
accidental reverse bias.)

Thus, I believe its probably not a good idea to put electrolytic
capacitors
back to back, as suggested previously.

Regards
Bob Monsen

This comes up about once every two or three years.

It *IS* OK to put two polarized caps back-to-back to make an AC
cap out of them. The reason is that there is a tiny bit more leakage
in the reverse biased cap. That leakage charges the "properly" biased
cap in the right direction to keep it happy when the current reverses.
Since the circuit is symetrical, each cap's leakage protects the
other.

Do a test with real capacitors (or even a SPICE simulation). You
will find that after only a couple of cycles that each cap will have a
DC voltage across it that is of a magnitude and polarity such that it
will be completely protected from reverse voltages.

You can guild the lilly with a couple of diodes. One diode
across each cap in a direction that will prevent the cap's voltage
from going more than a diode's drop reverse biased. The diodes are
not necessary but may give you that belt and suspenders feeling of
confidence.

Jim

 

[Thierry Nouspikel]

"Lewin A.R.W. Edwards" wrote:

{page write code}
// send CE# low, CLE high
// put Sequential Data Input command on data bus
// send WE# low
// bring WE# high
// bring CE# high

Aha! This is something I'm not doing. I keep CE# low all the time.

// bring CLE low
// bring CE# low

I'm gonna try this! I doubt that it is the problem, but who knows...

Thanks for your help,

Thierry

 

[DarkMatter]

On Wed, 23 Jul 2003 22:33:30 GMT, ben_nospam_bradley@mindspring.com
(Ben Bradley) Gave us:

Be careful with electronics, and take anything you read on Usenet
with a grain silo of salt.

That would be one high sodium intake!