About Us

Paul D. Spiegel

What do RATS use as a "rule of thumb" to select a fuse for a power amp?
In the past I've doubled the idle current, but honestly, that was just a
guess. How do you choose?

Phil Allison[_3_]

"Paul D. Spiegel"

What do RATS use as a "rule of thumb" to select a fuse for a power amp?
In the past I've doubled the idle current, but honestly, that was just a
guess. How do you choose?

** Kinda depends a lot a WHERE in the circuit the fuse is going to go.

Incoming AC supply ?

HT line before or after filter caps ?

Output tube cathodes ?

...... Phil

mick

On Mon, 30 May 2011 22:46:12 -0700, Paul D. Spiegel wrote:

What do RATS use as a "rule of thumb" to select a fuse for a power amp?
In the past I've doubled the idle current, but honestly, that was just a
guess. How do you choose?

You need to look at what the fuse is going to protect, and remember that
the fuse rating is, nominally, the maximum that the fuse will carry
continuously without blowing. Fuses blow because of heat, they have a
reverse time characteristic. As the current increases they blow faster.
That affects how you use them. It's no use putting a quick blow fuse in
part of the circuit that is going to see large surge currents, likewise a
slow blow fuse in the cathode of an output valve would never protect it.

To find out the actual blowing point of the fuse you need to get it's
fusing characteristics graph. That plots current against time, with a
tolerance included. From that, and knowing the load that you want to
protect, you can pick the correct fuse.

Or you can measure the load and make an intelligent guess ... :-)

--
Mick (Working in a M$-free zone!)
Web: http://www.nascom.info
Filtering everything posted from googlegroups to kill spam.

Tabby

On May 31, 6:59*am, "Phil Allison" wrote:
"Paul D. Spiegel"

What do RATS use as a "rule of thumb" to select a fuse for a power amp?
In the past I've doubled the idle current, but honestly, that was just a
guess. *How do you choose?

** Kinda depends a lot a WHERE *in the circuit the fuse is going to go.

very much so

Incoming AC supply ?

It also depends in this case on the supply voltage. Eg for 240v, for a
general purpose audio amp, 2A fuses tend not to last so well, 3A do.
On 110v you're likely not looking at a slightly fragile low current
value like 2A.

HT line before or after filter caps ?

which also depends on the resistance/impedance in teh circuit, since
that determines the power up surge. And reservoir cap size determines
its duration - eg a prewar amp with 4uF main resevoir cap will see way
less i squared r at startup than a later amp with maybe 100uF. Also
the type of rectifier: a valve rec will pretty much eliminate power up
surge, but silicon diodes wont.

Output tube cathodes ?

..... *Phil

NT

Patrick Turner

On May 31, 3:46*pm, "Paul D. Spiegel" wrote:
What do RATS use as a "rule of thumb" to select a fuse for a power amp? *
In the past I've doubled the idle current, but honestly, that was just a
guess. *How do you choose?

Fuses are a primitive way to stop a peice of gear smoking, and perhaps
burning your house down. Sometimes several fuses for an amp are wise
to stop parts from becoming damaged, such as OPTs, tubes etc.
Chosing the value is a bit tricky, but my rule of thumb is to use the
lowest possible fuse value which may indeed blow after say 200 hours
of normal use because of the heat from turn on cycles in the fuse will
fatigue the wire.

So for example, one estimates mains input current and selects a fuse
rated for slightly more and you try it. If it blows immediately on
turn on, try a slow blow type if you have not already, and maybe you
need bigger current value - repeat this until you find it lasts beyond
a few weeks. It will usually be a fuse value much smaller than that
supplied by a manufacturer who gets bad press if fuses blow too
easily. But I have had to rewire amps which have burnt to a crisp
because the mains fuse value was too high and the power tranny fried,
windings must have glowed red hot, sent flames and smoke out to the
rest of the amp.

Slow blow types give less "nuisance blows" because they'll tolerate a
brief high current, but will still fail if the current stays at the
rated value for more than say a few seconds. So 200mA fuses in the
cathode to 0V circuit of each output tube should be slow blow types. I
sometimes put fuses in series with all sec windings on a PT and on the
primary.

Patrick Turner.

Phil Allison[_3_]

"Patrick Turner"

Fuses are a primitive way to stop a peice of gear smoking, and perhaps
burning your house down.

** Fuses do not stop electronic gear from smoking and catching fire - the
number of fires from TV sets and CRT monitors is proof of that.

Fires do not take hundreds of watts to start and where the cabinetry is of
flammable plastic - watch out !!

So for example, one estimates mains input current ...

** Huh ??

Might be an idea to MEASURE it with a * true rms * AC current meter - allow
for the worst case operating conditions and then add say 20%.

Slow blow fuses are essential for AC supply fuses used with anything that
has an internal DC supply.

BTW:

A user replaceable fuse is no protection against ANYTHING !!!

..... Phil

Paul D. Spiegel

Phil Allison wrote:

"Paul D. Spiegel"

What do RATS use as a "rule of thumb" to select a fuse for a power amp?
In the past I've doubled the idle current, but honestly, that was just a
guess. How do you choose?

** Kinda depends a lot a WHERE in the circuit the fuse is going to go.

Incoming AC supply ?

HT line before or after filter caps ?

Output tube cathodes ?

..... Phil

Right. I should have been more precise. I am referring to the incoming
AC. This particular amp draws about 0.5a at idle so I would normally
choose a 1a fuse. I was curious how other RATS choose a fuse rating.

- Paul

Paul D. Spiegel

Tabby wrote:
On May 31, 6:59?am, "Phil Allison" wrote:
"Paul D. Spiegel"

What do RATS use as a "rule of thumb" to select a fuse for a power amp?
In the past I've doubled the idle current, but honestly, that was just a
guess. ?How do you choose?

** Kinda depends a lot a WHERE ?in the circuit the fuse is going to go.

It also depends in this case on the supply voltage. Eg for 240v, for a
general purpose audio amp, 2A fuses tend not to last so well, 3A do.
On 110v you're likely not looking at a slightly fragile low current
value like 2A.

HT line before or after filter caps ?

which also depends on the resistance/impedance in teh circuit, since
that determines the power up surge. And reservoir cap size determines
its duration - eg a prewar amp with 4uF main resevoir cap will see way
less i squared r at startup than a later amp with maybe 100uF. Also
the type of rectifier: a valve rec will pretty much eliminate power up
surge, but silicon diodes wont.

Interesting point, NT. In this case the line voltage is 120 (I forgot
to mention in my previous reply). The PS for this amp is a tube
rectifier with a choke input, so start-up current is minimized, but it
is something I should consider when using SS diodes and a large filter
cap first.

Phil Allison[_3_]

"Paul D. Spiegel"
"Phil Allison"
"Paul D. Spiegel"

What do RATS use as a "rule of thumb" to select a fuse for a power
amp?
In the past I've doubled the idle current, but honestly, that was just
a
guess. ?How do you choose?

** Kinda depends a lot a WHERE ?in the circuit the fuse is going to go.

In this case the line voltage is 120 (I forgot
to mention in my previous reply). The PS for this amp is a tube
rectifier with a choke input, so start-up current is minimized,

** The AC tranny has a huge magnetic inrush surge.

10 to 50 times the running current.

Fool.

.... Phil

Patrick Turner

On Jun 1, 1:33*pm, "Phil Allison" wrote:
"Patrick Turner"

Fuses are a primitive way to stop a peice of gear smoking, and perhaps
burning your house down.

** Fuses do not stop electronic gear from smoking and catching fire - *the
number of fires from TV sets and CRT monitors is proof of that.

I have had hundreds of items pass over my workbench for a repair. Some
had evidence of small fires, and one EAR509 amp had suffered a mans
tranny which had spewed flames and smoke, damaging the rest of the
amp. But I don't get to see the melted mess which may result from a TV
or PC monitor fire.

I did mention that the fuse current value should usually be lower than
what the manufacturer provides. It may not stop ALL fires, but it will
stop many more than if the fuse value is too high.

Many old radio chassis in tinder dry wooden cabinets are a worry, so I
always add a mains fuse during service work because very few old
radios had a fuse.

Fires do not take hundreds of watts to start and where the cabinetry is of
flammable plastic - watch out *!!

Indeed.

So for example, one estimates mains input current ...

** Huh ??

I beleive I was talking to someone not very technically qualified, and
he may not own a meter, and like so many, could not use a meter to do
anything at all. A significant number of people have SFA idea about
volts, amps, watts, ohms, or electrons etc.

So, by saying one estimates input current, I should have said * I *
might estimate current, add up the HT x Iadc, heater powerr, add 20%
for losses, get a total in VA, divide by 240, and you have input
current, which will not be a pure sine wave. OK, then you try some
slow blow fuses until you find a value that will last a few weeks at
least.

But in all my amps and re-engieered amp I put in active circuits to
detect excessive Idc in output tubes which is the main cause of
serious amp troubles, ie bias failures. Many have thanked me for it
because the problem turns off the amp, and they bring it to me for a
fix which is easy and usually done in 30 mins, often one output tube
has lost its ability to stay biased and Iadc escalates.

Might be an idea to MEASURE it with a * true rms * AC current meter - allow
for the worst case operating conditions and then add say 20%.

Indeed.

Slow blow fuses are essential for AC supply fuses used with anything that
has an internal DC supply.

BTW:

A user replaceable fuse is no protection against ANYTHING !!!

Some of my amps have 6 fuses, all mounted in a row in screw down
holders and accessable by an owner. And beside each is a red LED that
turns on to tell you which fuse is blown. They are in series with
every transformer winding used in the amp. In addition, there is an
active circuit which turns the amp off at the mains if any one or more
of many OP tubes is in bias trouble. I don't like getting sued. I
don't want to have to rewind any OPTs. So there is protection for
about anything i can think of.

Patrick Turner.

.... *Phil

Big Bad Bob

On 05/31/11 23:43, Phil Allison so wittily quipped:

In this case the line voltage is 120 (I forgot
to mention in my previous reply). The PS for this amp is a tube
rectifier with a choke input, so start-up current is minimized,

** The AC tranny has a huge magnetic inrush surge.

10 to 50 times the running current.

Fool.

NO, it will not be that high, though there is likely to be SOME surge
for _MAYBE_ 1 cycle (depending on where the current starts flowing.
Once again, you're an idiot. I only use the ad-hominem attack because
you seem to want to elevate yourself at others' expense, and that's most
likely due to your lack of real understanding. Obviously you failed
trigonometry, at the very least, or you would understand the effect of a
sine wave AC voltage on an inductor. However, transformers aren't pure
inductors, though an inductive component is likely to be there.

In any case, it's a transformer, not a DC electromagnet or a motor. You
don't have to 'magnetize' it. There is no 'magnetic inrush surge'.
That's just ridiculous. There is only the effective inductance and the
reflected impedence of the load.

Now as for surge currents, they DO exist (obvious to the most casual
observer with any reasonable common sense level of understanding of
electronics and electricity) but not for the reasons you stated. It's
more like charging up capacitors and low resistance on cold tube filaments.

Go back to school and learn what you're talking about BEFORE you call
someone 'fool'. Fool.

/me ignoring whatever reply you give - it won't be worth responding to.
just the usual ad-hominem attacks due to lack of real substance,
attempting to elevate yourself above the rest of us through typical Jr.
High bullying, yotta yotta yotta boring as hell. This tirade has been
for the benefit of those you insult, not you. Saying this to you is a
waste of time, of course, if it were intended for YOUR benefit. The
classic 'pearls before swine' example applyies.

Patrick Turner

On Jun 1, 4:43*pm, "Phil Allison" wrote:
"Paul D. Spiegel"
*"Phil Allison"

"Paul D. Spiegel"

What do RATS use as a "rule of thumb" to select a fuse for a power
amp?
In the past I've doubled the idle current, but honestly, that was just
a
guess. ?How do you choose?

** Kinda depends a lot a WHERE ?in the circuit the fuse is going to go.

*In this case the line voltage is 120 (I forgot
to mention in my previous reply). *The PS for this amp is a tube
rectifier with a choke input, so start-up current is minimized,

** The AC tranny has a huge magnetic inrush surge.

* *10 to 50 times the running current.

* *Fool.

*.... *Phil

Here we go again Phil. This topic of inrush current to magnetize the
core has been discussed before at length. But if working RMS current
is 0.5Arms, I've never seen a 5 amp peak let alone a 25A peak which
*might* require a fuse rated for such currents. Typical mains fuse for
Quad-II with completely revised PSU using input filter of
470uF-150r-470uF and 4 x 3A 1N5408 diodes can be 1A. No evidence of
needing any more, although Quad used 2A for 240V. The main prolonged
inrush current is during the first 2 seconds to charge up all that
capacitance, But I limit charge currents with series R between ends of
HT winding and diodes. With such R and the high PT winding resistance
in Quad-II PT, charge currents are limited.
Even in amps with 200,000 uF and +/-70V rails, and very low resistance
windings, the magnetic input current seems less of a problem than that
of charging up the caps. So I add an R in series with mains, and when B
+, B- get up to 0.75 full value, the R is shunted via a relay and
delay circuit powered by the turn on management and protection
auxilliray supply. The initial peak is limited, and the second peak is
usually less than the first, and usually the mains fuse may be halved
in value compared to what is needed without the inrush current
limiter.

Patrick Turner

Big Bad Bob

On 05/30/11 22:46, Paul D. Spiegel so wittily quipped:
What do RATS use as a "rule of thumb" to select a fuse for a power amp?
In the past I've doubled the idle current, but honestly, that was just a
guess. How do you choose?

doubling idle current is probably not big enough.

I typically pick a fuse size that is small enough that it won't blow
under max load conditions, but WILL blow if something fries, _before_
other things get permanent damage (or catch on fire). Input fuses are
likely to be slow blow, and fast blow for DC power supply fuses (if you
are able to limit surge current on capacitor charging).

For the AC input, I'd use a slow blow fuse and pick a value that's twice
'maximum expected current + starting surge' as a good starting point and
adjust as needed (the lower you can get away with, the better). It
should also blow fast enough so that if you get a short, it prevents
fires and electrocution.

anyway you can google on how fuses are rated for more info.

Phil Allison[_3_]

"Patrick Turneroid the Bricklayer "

Fuses are a primitive way to stop a peice of gear smoking, and perhaps
burning your house down.

** Fuses do not stop electronic gear from smoking and catching fire - the
number of fires from TV sets and CRT monitors is proof of that.

( snip usual Turneroid absurd off-topic DRIVEL )

Fires do not take hundreds of watts to start and where the cabinetry is of
flammable plastic - watch out !!

So for example, one estimates mains input current ...

** Huh ??

Might be an idea to MEASURE it with a * true rms * AC current meter -
allow
for the worst case operating conditions and then add say 20%.

Indeed.

Slow blow fuses are essential for AC supply fuses used with anything that
has an internal DC supply.

BTW:

A user replaceable fuse is no protection against ANYTHING !!!

Some of my amps have 6 fuses,

** Some of you feet have six toes.

...... Phil

Phil Allison[_3_]

"Patrick Turneroid the Bricklayer "

** The AC tranny has a huge magnetic inrush surge.

10 to 50 times the running current.

Fool.

Here we go again Phil.

** **** off - you know nothing RDH4 freakoid.

This topic of inrush current to magnetize the
core has been discussed before at length.

** Then how come YOU never learned anything then ??

Head too far stuck up your puny arse ??

But if working RMS current
is 0.5Arms, I've never seen a 5 amp peak let alone a 25A peak which
*might* require a fuse rated for such currents.

** Fraid that is how it is all the time.

Slow blow fuses are all rated to withstand surges of 10 to 20 times I
nominal for that reason.

You illiterate ****head.

..... Phil

Big Bad Bob

On 06/01/11 03:51, flipper so wittily quipped:
On Wed, 01 Jun 2011 02:19:46 -0700, Big Bad Bob
wrote:

On 05/31/11 23:43, Phil Allison so wittily quipped:

In this case the line voltage is 120 (I forgot
to mention in my previous reply). The PS for this amp is a tube
rectifier with a choke input, so start-up current is minimized,

** The AC tranny has a huge magnetic inrush surge.

10 to 50 times the running current.

Fool.

NO, it will not be that high, though there is likely to be SOME surge
for _MAYBE_ 1 cycle (depending on where the current starts flowing.
Once again, you're an idiot. I only use the ad-hominem attack because
you seem to want to elevate yourself at others' expense, and that's most
likely due to your lack of real understanding. Obviously you failed
trigonometry, at the very least, or you would understand the effect of a
sine wave AC voltage on an inductor. However, transformers aren't pure
inductors, though an inductive component is likely to be there.

Likely?

In any case, it's a transformer, not a DC electromagnet or a motor. You
don't have to 'magnetize' it. There is no 'magnetic inrush surge'.
That's just ridiculous. There is only the effective inductance and the
reflected impedence of the load.

Not so.

http://www.opamp-electronics.com/tut...nt_2_09_12.htm

uh, that's not the same thing. I already mentioned this kind of thing,
actually. That 'surge' typically happens within 1/2 a cycle, and is
significanly less than 1 cycle. But any current spike would be so small
as to be insignificant when compared to actual LOAD CURRENT. Sure it
might show up on an o-scope, but events that are less than one cycle in
length aren't going to blow fuses. In fact, it would be HEAVILY damped
by the inductance of the transformer. Remember, an inductor starts
conducting with ~ahem~ MINIMUM current (highest counter-EMF) which then
drops as the inductor 'charges', as it's behavior is effectively the
opposite of a capacitor. But it's not likely to exceed its steady-state
reactive current, if at all. That is also minimal when compared to the
actual load current. To test you could connect an o-scope to a CT then
plug something in that's already switched on, and see what the surge
looks like. Might be fun to actually do it, eh?

A 1H inductor increases its current at 1A per sec when 1V is applied
(it's a linear increase). For 120V, that's 160A/sec (worst case,
starting at the peak). 1/60th of a sec means no more than 3A (worst
case, unlikely though, due to Z). Impedence at 60Hz would be around 400
ohms (2piFL), so peak reactive current is going to be around 0.3A (much
smaller than the 3A potential surge). We know that once current
stabilizes it will asymptotically approach its maximum, lagging the
voltage by 90 degrees. And of course a transformer primary is very
likely to be more than 1H. Suffice it to say that surge current due to
the transformer itself is very, very, very insignificant. Now with a
resistive load on the secondary that draws 10 times the current (tube
heaters) I can believe there will be a surge. But that's not from the
transformer. That's the reflected load at ~10 times normal.

Phil Allison[_3_]

"Big Bad Bob"

** This ****ing **** needs a bullet in his head.

Big Bad Bob

On 06/01/11 02:22, Patrick Turner so wittily quipped:
Here we go again Phil. This topic of inrush current to magnetize the
core has been discussed before at length.

only because there appear to be some definitions for it that don't
include audio amplifier power supply transformers. In some cases,
capacitive loads exist, and in other cases, very high current cores.
Undoubtedly the 'inrush' is pale by comparison to the typical LOAD
CURRENT. And that's my point. it's effectly non-existent except for
special cases, in which the current surge is dampened by the inductance
of the transformer, and DOES NOT EXCEED a fraction of the load-carrying
capacity of the device.

But if working RMS current is 0.5Arms, I've never seen a 5 amp peak
let alone a 25A peak which *might* require a fuse rated for such currents.

Your observations are correct. To say there is a surge of 50 times
normal operating current is just, plain, ignorant (not counting what may
happen within the 1st cycle of closing the switch). That of course does
not count if significant arcing happens, internally or otherwise. If
your transformer turns into an AC arc welder, all bets are off. That's
a much different scenario.

But I limit charge currents with series R between ends of
HT winding and diodes. With such R and the high PT winding resistance
in Quad-II PT, charge currents are limited.

yes (good design). our 'expert' should re-read the pages for the links
he posted. Alternate design would use an LC for the same reason, and
better filtering. L also provides some series R for charge damping, and
nice inductive action to make the current waveform a bit more sinusoidal.

ever measure startup current with an o-scope across a small R or with a
CT? just curious.

Phil Allison[_3_]

"Big Bad Boob = TOTAL ****HEAD "

FFS see:

http://en.wikipedia.org/wiki/Inrush_current

YOU vile, retarded pile of ****ing sub human dung.

..... Phil

Big Bad Bob

On 06/01/11 02:05, Patrick Turner so wittily quipped:
Some of my amps have 6 fuses, all mounted in a row in screw down
holders and accessable by an owner. And beside each is a red LED that
turns on to tell you which fuse is blown.

Ne bulbs (for HV circuits) work well also, but probably cost more. A
lot of fuse holders the US Navy uses are equipped with Ne bulbs and
resistors across the fuse as 'blown fuse' indicators. Helps
troubleshoot problems quickly. "Hey, there's a blown fuse"

extra protection helps prevent costly repairs later on when someone
throws a ball for the dog and it bounces off his nose and breaks one of
the output tubes while the unit is running at power. Or when someone
hits a golf ball through your window and into your amp. oops.

(never happens, right?)

so it's nice to see you put that kind of defensive design into these things.

GRe

"Paul D. Spiegel" wrote in message
...

Right. I should have been more precise. I am referring to the incoming
AC. This particular amp draws about 0.5a at idle so I would normally
choose a 1a fuse. I was curious how other RATS choose a fuse rating.

- Paul

Like others said, it depends, you must know (measure the current)
instantaneous current behaviour the first seconds after power on.

I've placed an example I-t curves .pdf on A.B.S.E. that gives an
idea of current stress behaviour of (particular/0.5A) fuses.
Data originates from Little-Fuse, the curves are NOT generic.

Gio

Phil Allison[_3_]

"flipper"

But any current spike would be so small
as to be insignificant when compared to actual LOAD CURRENT.

It would seem you didn't read the page. With an 'ideal transformer',
which none are,

** An ideal transformer does not suffer from saturation at all.

So the issue of " inrush surge " simply does not exist.

Wot absurd, ****ing sophistry.

.... Phil

Phil Allison[_3_]

"flipper the ****head "

** An ideal transformer does not suffer from saturation at all.

So the issue of " inrush surge " simply does not exist.

Yes it does, at twice operating current.

** What is the " operating current " of an unloaded, ideal transformer that
by definition does not have any magnetising current or ever saturate ??

Answer: Nothing.

The term "inrush surge " only applies to REAL transformers that DO saturate.

You ****ing, stupid, autistic ****head.

..... Phil

Phil Allison[_3_]

flipper the ****ing ****head "

** An ideal transformer does not suffer from saturation at all.

So the issue of " inrush surge " simply does not exist.

Yes it does, at twice operating current.

** What is the " operating current " of an unloaded, ideal transformer that
by definition does not have any magnetising current or ever saturate ??

Answer: Nothing.

The term "inrush surge " only applies to REAL transformers that DO saturate.

You ****ing, stupid, autistic ****head.

That stupid link was 100% WRONG.

..... Phil

Big Bad Bob

On 06/02/11 05:52, flipper so wittily quipped:
On Thu, 02 Jun 2011 02:44:27 -0700, Big Bad Bob
wrote:

On 06/01/11 03:51, flipper so wittily quipped:
On Wed, 01 Jun 2011 02:19:46 -0700, Big Bad Bob
wrote:

On 05/31/11 23:43, Phil Allison so wittily quipped:

In this case the line voltage is 120 (I forgot
to mention in my previous reply). The PS for this amp is a tube
rectifier with a choke input, so start-up current is minimized,

** The AC tranny has a huge magnetic inrush surge.

10 to 50 times the running current.

Fool.

NO, it will not be that high, though there is likely to be SOME surge
for _MAYBE_ 1 cycle (depending on where the current starts flowing.
Once again, you're an idiot. I only use the ad-hominem attack because
you seem to want to elevate yourself at others' expense, and that's most
likely due to your lack of real understanding. Obviously you failed
trigonometry, at the very least, or you would understand the effect of a
sine wave AC voltage on an inductor. However, transformers aren't pure
inductors, though an inductive component is likely to be there.

Likely?

In any case, it's a transformer, not a DC electromagnet or a motor. You
don't have to 'magnetize' it. There is no 'magnetic inrush surge'.
That's just ridiculous. There is only the effective inductance and the
reflected impedence of the load.

Not so.

http://www.opamp-electronics.com/tut...nt_2_09_12.htm

uh, that's not the same thing.

Not 'the same thing' as WHAT?

The surge is due to the magnetic flux so calling that a 'magnetic
inrush surge' is appropriate.

It would seem you didn't read the page.

your "sort of right", I actually 'skimmed' the page, saw what it was
saying (and the basis of it), didn't waste time on details reading EVERY
GODDAMNED WORD ON IT that I didn't see as being important. It was
obvious what they were saying, and I agree with the general concept, but
do not draw the same conclusions from it. (saying I did not read
something when I simply disagree is pointless).

I'll explain it differently now. You get a slight 'surge' when you snap
the switch at a particular 'non-ideal' point in the incoming voltage's
AC cycle [already mentioned]. Magnetic hysteresis is NOT that large for
a normal power transformer (typically don't use "bar iron" for that
right?), so the transformer will behave MOSTLY as an inductor (at the
beginning). You could (in theory) construct a device that DOES have
'surges' but why would you want to? The 'surge' will disappear within 1
cycle, and won't be larger than the normal current carrying capacity of
the device. Remember, the 'magnetization' process creates COUNTER EMF
(repeat COUNTER EMF) which is how inductance works. This magnetization
constantly switches polarity for AC, so even if the 'last magnetic
state' is in opposition to the current state (worst case) the surge
still won't be that large because of (drum roll please) the COUNTER EMF
generated by the magnetization reversal!

It's the same magnetic field changes that induce current in the
secondary. While the transformer is allegedly 'surging' the induced
current in the secondary won't be quite as much as it is while it's
'running'. But, again, less than 1 cycle. 1/60th of a second is too
short of a time to overload even a FAST BLOW fuse. Fuse blow times are
significantly longer than 1 cycle unless current is extremely high (like
a dead short) because you need to heat the conductor before it melts.
You would get more of a surge from charging capacitors with a silicon
rectifier power supply than from the transformer 'building a magnetic
field' (which COMPLETELY reverses 120 times per second for 60hz AC, FYI).

Remember how motors and generators work. A changing magnetic field in a
wire creates EMF. Although hysteresis affects the amount of current
needed to change the magnetic field, hysteresis in transformers (by
design, unless they're CRAPPY) is VERY VERY LOW. Otherwise, it would
overheat in normal operation, and we don't want THAT. And of course,
I'm not talking about 'ideal transformers' (which would have NO surge).

With an 'ideal transformer',
which none are, the inrush current due to the magnetic flux would be
exactly *twice* the operating current.peak which, even alone, is not
'insignificant'.

No, it would NOT be 'twice the operating current peak'. That's silly.
"idle current" peak I would accept (with the transformer unloaded) for
the reasons I mentioned above. But transformer idle current is
extremely small, or else the transformer would (again) overheat. Last
time I checked a transformer that had been energized with no load on it
for a significant period of time, there was no significant heat coming
from it. However, same transformer with a load, nice and toasty.
Conclusion: current losses in the transformer increase under load,
suggesting that total primary winding current is insignificant when
unloaded, significant when loaded. I think that's nice logical common
sense, and "not ideal" as well as "not pure theory".

I suspect you're using pure theory in your analysis. But you might be
misinterpreting it a bit. I suggest a simple experiment, measuring the
actual current spike. I'll trust your results, so I don't have to do it
myself. And I won't accuse you or anything if you decide NOT to do it,
rather than having accusations tossed at me for allegedly "not reading".
Now I've been known to throw that sort of thing back into the face of
people that (typically?) throw it at me, but I'm trying NOT to 'go
there' today. It's more interesting to figure out what's really
happening when you power up a transformer and not get into a ****ing
contest (I'll reserve that sort of thing for the 1 or 2 trolls that are
addictively compelled to invent new profanities every day to describe me).

anyway I'd put a slow blow fuse on the input because of capacitor
charging and heater warmup, NOT transformer magnetization.

Big Bad Bob

On 06/02/11 09:07, flipper so wittily quipped:
On Fri, 3 Jun 2011 00:30:54 +1000, "Phil
wrote:

flipper the ****ing ****head "

** An ideal transformer does not suffer from saturation at all.

So the issue of " inrush surge " simply does not exist.

Yes it does, at twice operating current.

** What is the " operating current " of an unloaded, ideal transformer that
by definition does not have any magnetising current or ever saturate ??

Answer: Nothing.

The term "inrush surge " only applies to REAL transformers that DO saturate.

You ****ing, stupid, autistic ****head.

That stupid link was 100% WRONG.

Now you're a babbling fool asshole.

he's trolling you. You _could_ troll him back, and point out that he'd
argue with a math teacher that tells him that 1 + 1 = 2, just because he
argues with everyone and calls them '****head'. Actually, if he's going
to insult someone, I think "vacuous toffee-nosed malodorous pervert"
works better than "****ing stupid autistic ****head" simply because the
former is eloquent and funny as hell (especially when spoken by Graham
Chapman), and the latter only requires a room temperature IQ (measured
in Celcius).

Big Bad Bob

On 06/02/11 06:15, flipper so wittily quipped:
On Wed, 1 Jun 2011 02:22:19 -0700 (PDT), Patrick Turner
wrote:

On Jun 1, 4:43 pm, "Phil wrote:
"Paul D. Spiegel"
"Phil Allison"

"Paul D. Spiegel"

What do RATS use as a "rule of thumb" to select a fuse for a power
amp?
In the past I've doubled the idle current, but honestly, that was just
a
guess. ?How do you choose?

** Kinda depends a lot a WHERE ?in the circuit the fuse is going to go.

In this case the line voltage is 120 (I forgot
to mention in my previous reply). The PS for this amp is a tube
rectifier with a choke input, so start-up current is minimized,

** The AC tranny has a huge magnetic inrush surge.

10 to 50 times the running current.

Fool.

.... Phil

Here we go again Phil. This topic of inrush current to magnetize the
core has been discussed before at length. But if working RMS current
is 0.5Arms, I've never seen a 5 amp peak let alone a 25A peak

You most likely wouldn't under casual observation because it only
occurs if AC is applied during, or close to, zero crossing and the
odds of you just randomly hitting it with the power switch are low.

Sooner or later it will happen, though.

which
*might* require a fuse rated for such currents.

Called 'slo-blo'.

Minimum blow time for a 1A slo-blo at 40 amp surge is 10mS.

heh, half a cycle at 50Hz. Lucky us that won't happen without a serious
short somewhere.

I think 'slo-blo' was a registered trademark (several sites say
Littlefuse Inc owns it, didn't research further, see link). It's
probably like calling facial tissues "kleenex" or super-sticky utility
tape "duck tape", becoming an accepted 'standard' term.

trademark reference
http://trade.mar.cx/US72009368

Big Bad Bob

On 06/02/11 12:45, flipper so wittily quipped:
With an 'ideal transformer',
which none are, the inrush current due to the magnetic flux would be
exactly *twice* the operating current.peak which, even alone, is not
'insignificant'.

No, it would NOT be 'twice the operating current peak'.

Yes, it will. Read the damn page.

no. my mind will not be changed without an experiment first.

Big Bad Bob

On 06/02/11 12:51, flipper so wittily quipped:
On Thu, 02 Jun 2011 12:36:44 -0700, Big Bad Bob
wrote:

On 06/02/11 06:15, flipper so wittily quipped:
On Wed, 1 Jun 2011 02:22:19 -0700 (PDT), Patrick Turner
wrote:

On Jun 1, 4:43 pm, "Phil wrote:
"Paul D. Spiegel"
"Phil Allison"

"Paul D. Spiegel"

What do RATS use as a "rule of thumb" to select a fuse for a power
amp?
In the past I've doubled the idle current, but honestly, that was just
a
guess. ?How do you choose?

** Kinda depends a lot a WHERE ?in the circuit the fuse is going to go.

In this case the line voltage is 120 (I forgot
to mention in my previous reply). The PS for this amp is a tube
rectifier with a choke input, so start-up current is minimized,

** The AC tranny has a huge magnetic inrush surge.

10 to 50 times the running current.

Fool.

.... Phil

Here we go again Phil. This topic of inrush current to magnetize the
core has been discussed before at length. But if working RMS current
is 0.5Arms, I've never seen a 5 amp peak let alone a 25A peak

You most likely wouldn't under casual observation because it only
occurs if AC is applied during, or close to, zero crossing and the
odds of you just randomly hitting it with the power switch are low.

Sooner or later it will happen, though.

which
*might* require a fuse rated for such currents.

Called 'slo-blo'.

Minimum blow time for a 1A slo-blo at 40 amp surge is 10mS.

heh, half a cycle at 50Hz. Lucky us that won't happen without a serious
short somewhere.

You'll certainly never know because you never bother reading what's
written.

I can't believe you'd actually say something like that, especially when
you (hypocritically) failed to actually do the EXPERIMENT I mentioned.
Some people, and their arrogance. I thought you were ABOVE that. I
guess not.

Big Bad Bob

On 06/02/11 13:51, flipper so wittily quipped:
On Thu, 02 Jun 2011 12:47:53 -0700, Big Bad Bob
wrote:

On 06/02/11 12:45, flipper so wittily quipped:
With an 'ideal transformer',
which none are, the inrush current due to the magnetic flux would be
exactly *twice* the operating current.peak which, even alone, is not
'insignificant'.

No, it would NOT be 'twice the operating current peak'.

Yes, it will. Read the damn page.

no. my mind will not be changed without an experiment first.

Plenty of people have done it and it's common knowledge in the power
industry so there's no reason for me to waste my time recreating the
wheel. Especially when it's no skin off my nose.

http://www.aedie.org/9CHLIE-paper-send/273_manana.pdf

http://www.apqi.org/file/attachment/2009424/162650.pdf

But you probably won't read that either.

true, it's getting boring now and I have other things to do. And you
completely missed my point ('disagreeing' is not the same thing as 'not
reading'). Or, just because someone on the intarweb writes something
down, it does not imply that it's either ACCURATE or repeatable through
experimentation. Real science questions every theory and repeats
experiments all of the time. 'Re-inventing the wheel' happens every
year in Detroit and in the video game industry, out of necessity I might
add, and it's never a BAD thing to discover better ways of making wheels
[or anything else]. That's sometimes called ENGINEERING. In any case,
I'm sure you're arguing details that I consider unimportant, so I'll
stop now. I'm sure it's pointless to continue, since you're not going
to change my mind, and I doubt I'll change yours.

There are people out there who really hate it when someone disagrees
with them. Too bad.

I'll stop now, it's obviously pointless to continue.

Fred[_13_]

"Big Bad Bob" wrote in message m...
On 06/02/11 13:51, flipper so wittily quipped:

snip

true, it's getting boring now and I have other things to do. And you completely missed my point ('disagreeing' is not the same
thing as 'not reading'). Or, just because someone on the intarweb writes something down, it does not imply that it's either
ACCURATE or repeatable through experimentation. Real science questions every theory and repeats experiments all of the time.
'Re-inventing the wheel' happens every year in Detroit and in the video game industry, out of necessity I might add, and it's
never a BAD thing to discover better ways of making wheels [or anything else]. That's sometimes called ENGINEERING. In any case,
I'm sure you're arguing details that I consider unimportant, so I'll stop now. I'm sure it's pointless to continue, since you're
not going to change my mind, and I doubt I'll change yours.

Dude!

The huge current surge in a power xfmr when power is applied at the zero-crossing
voltage point isn't a matter of opinion, it's a well known phenomenon that has been
well known for as long as they've been winding transformers! You'd know that if
you read the literature, but you, well, don't.

Get yourself a cheap zero-crossing solid state relay from All Electronics and find
out for yourself if you won't read or refuse to believe any of the engineering texts
out there.

Fred

There are people out there who really hate it when someone disagrees with them. Too bad.

I'll stop now, it's obviously pointless to continue.

Big Bad Bob

On 06/02/11 23:15, flipper so wittily quipped:
On Thu, 02 Jun 2011 16:19:09 -0700, Big Bad Bob
wrote:

On 06/02/11 12:51, flipper so wittily quipped:
On Thu, 02 Jun 2011 12:36:44 -0700, Big Bad Bob
wrote:

On 06/02/11 06:15, flipper so wittily quipped:
On Wed, 1 Jun 2011 02:22:19 -0700 (PDT), Patrick Turner
wrote:

On Jun 1, 4:43 pm, "Phil wrote:
"Paul D. Spiegel"
"Phil Allison"

"Paul D. Spiegel"

What do RATS use as a "rule of thumb" to select a fuse for a power
amp?
In the past I've doubled the idle current, but honestly, that was just
a
guess. ?How do you choose?

** Kinda depends a lot a WHERE ?in the circuit the fuse is going to go.

In this case the line voltage is 120 (I forgot
to mention in my previous reply). The PS for this amp is a tube
rectifier with a choke input, so start-up current is minimized,

** The AC tranny has a huge magnetic inrush surge.

10 to 50 times the running current.

Fool.

.... Phil

Here we go again Phil. This topic of inrush current to magnetize the
core has been discussed before at length. But if working RMS current
is 0.5Arms, I've never seen a 5 amp peak let alone a 25A peak

You most likely wouldn't under casual observation because it only
occurs if AC is applied during, or close to, zero crossing and the
odds of you just randomly hitting it with the power switch are low.

Sooner or later it will happen, though.

which
*might* require a fuse rated for such currents.

Called 'slo-blo'.

Minimum blow time for a 1A slo-blo at 40 amp surge is 10mS.

heh, half a cycle at 50Hz. Lucky us that won't happen without a serious
short somewhere.

You'll certainly never know because you never bother reading what's
written.

I can't believe you'd actually say something like that,

Why? Because it's true?

no, it is not. you appear to have a lot of misconceptions and bad
assumptions.

OK I ran the experiment myself. Test circuit used 120VAC to 24VCT
transformer (rated at 2A) with 10 ohm resistor in series with the
primary. I could have used a scope and captured the actual waveform,
but instead I decided to focus on a specific point, that is the peak
surge current on power-up, so I put a diode and capacitor across it so I
could easily measure the peak (cap is 3.3uF 50V). After power cycling
dozens of times, the MAXIMUM value I ever read was 15V. I will assume
that is the worst case peak. I achieved similar voltages (let's say
12V) several times, but the 15V value only once. This suggests that the
peak 'inrush' current was 1.5A. Since the transformer is rated at
around 50VA we can assume that the actual 'inrush' current (in the worst
case measured case) was about 4 times the peak rating of the
transformer. the likelihodd (had I bothered to hook up a scope, I'll
let someone ELSE do that one, hah) is that it was less than 1/2 a cycle.
Putting a resistive load on the secondary did not appear to affect any
spikes. Also, I had no way of determining whether any arcing took place
within the switch during the measurement (which would have greatly
affected the test, causing the reading to be significantly higher).

So in retrospect, the instantaneous (measured) current flow was 1.5A in
the worst case, with ~400ma expected at full load. Assuming it's less
than 1 cycle in length (less than 10 msecs) no fast-blow fuse rated at
0.5A would have blown due to 'current inrush'. Further, it wasn't the
10 to 50 times the rated current as claimed by some. It was barely 4
times, and only for one measurement. An o-scope would give you a better
perspective on how long the surge lasts and the overall curve shape, but
like I said I didn't bother hooking one up (I'll let one of you guys do
it). And of course I never said it wouldn't be there, I said it would
be SMALL, particularly WAY LESS than the 'claimed' value. Most of the
measurements did not exceed 500ma. I didn't bother writing them down, I
simply watched the meter. But if it turns out that arcing was THE
primary cause for the occasional LARGE readings, then none of what
anyone ELSE said about these things will matter, because arcing throws
the measurements off.

(I would have used a 1 ohm resistor but I only had a 10 ohm high wattage
resistor handy).

let anyone else refute my experimental data and I will gladly accept the
results. But argue about theory and accuse me of "not reading" simply
because I disagree? Only repeatable experimental data is meaningful in
science. Everything else is theory, debatable, and since anyone can say
anything at all and claim it to be true, absolute bull**** until proven
experimentally.

Tabby

On Jun 3, 4:01*am, "Phil Allison" wrote:
"Tabby"
*"Phil Allison"

** The AC tranny has a huge magnetic inrush surge.

10 to 50 times the running current.

50x run current would seem to suggest a transformer regulation of 2 or
3%.

** Only true is the "running current " is such that is consumes the entire
VA rating of the tranny.

Cant say I've seen many of those in domestic valve amps. How do
you get the 50x figure?

** Say the tranny is a 300VA toroidal and the AC supply is 230 volts.

The worst case, half cycle peak surge is equal to the AC peak voltage
divided by the primary resistance - *which is typically 4.5 ohms. That gives
72 amps peak, worst case.

The running current can be no more than 1.3 amps rms ( gives 300VA) and is
more likely under 1 amp.

BTW:

If the first half cycle of magnetic surge current is positive, then all
subsequent half cycles are also positive decaying exponentially for 5 to 20
cycles until the core de-saturates.

Amazingly, the peaks of these half waves coincide closely with AC supply
zero crossings.

Any surge current due to the filter caps in the PSU charging add on top of
the magnetic surge and are of both polarities - the peaks in this case
coincide with AC supply voltage peaks.

The combined current wave is a treat to study and it is easy to see why slow
blow fuses are essential and larger transformers require a soft start system
in the AC supply feed.

.... *Phil

Interesting. If this is the case, presumably the same problem occurs
with inductors. I don't ever recall considering that in design, I'm
not big on inductors though, there are often better ways to do things.

NT

Phil Allison[_3_]

"Big Bad Bob"

OK I ran the experiment myself. Test circuit used 120VAC to 24VCT
transformer (rated at 2A) with 10 ohm resistor in series with the primary.

** Change that resistor to 1 ohm and try again.

Cos you went and put a resistor in series with the primary of similar value
to the primary's own resistance - which has the effect of dramatically
reducing the peak current of a worst case surge.

Folk regularly use NTC thermistors for that exact purpose !!!!!

Then try more transformers - particularly those of a few hundred VA and
toroidal types.

You will need a resistor about 0.1 to 0.5 ohms with plenty of power rating.

****head.

..... Phil

Phil Allison[_3_]