Power rating for 1-ohm bias sensing resistor

[PharmRock]

On Doug's site about adding the 1-ohm resistors between pin 8 and ground, he states "Now solder one end of a one ohm three watt [emphasis added] resistor to pin eight. Solder the other end to the wire that is soldered to the chassis." https://el34world.com/charts/fenderservice5.htm.

Why is a 3-watt resistor suggested here?  If we are setting the bias around 30-35mA, then we are looking at maybe 1 milliwatt of dissipation, correct?
P = I^2 x R = (0.03)^2 x 1 = 0.0009 watts.

Any reason I couldn't use a 1/2 watt 1% tolerance metal film?

[acheld]

I don't know why a 3 watt resistor was called for . . .

Hoffman normally specs a 1/2 watt metal film resistor for this purpose in his kit BOMs for Fender style amps.

[pdf64]

I think 1/2W is fine.  Bypass it with a diode, eg 1N400x. That will limit dissipation to 0.49W (ie 0.7 x 0.7/1), eg in the case of a valve shorting.
Beefy resistors there can roast nearby stuff https://www.thegearpage.net/board/index.php?threads/6g3-brown-deluxe-build-done-final-pix.2305389/#post-33586721

[Latole]

I always use 1/2 watt it ia way too powerfull

If you do hot bias, let say, 60 mv / mw reeding on a 1 ohms
60 X 60 = 0.00036 watts

[PharmRock]

I think 1/2W is fine.  Bypass it with a diode, eg 1N400x. That will limit dissipation to 0.49W (ie 0.7 x 0.7/1), eg in the case of a valve shorting.
Beefy resistors there can roast nearby stuff https://www.thegearpage.net/board/index.php?threads/6g3-brown-deluxe-build-done-final-pix.2305389/#post-33586721

sounds like a good backup safety measure. 
cathode of diode towards ground, correct?

[mresistor]

You don't need a diode - if a tube shorts and the 1 ohm resistor opens  that is a good thing.   

[EL34]

On Doug's site about adding the 1-ohm resistors between pin 8 and ground, he states "Now solder one end of a one ohm three watt [emphasis added] resistor to pin eight. Solder the other end to the wire that is soldered to the chassis." https://el34world.com/charts/fenderservice5.htm.

Why is a 3-watt resistor suggested here?  If we are setting the bias around 30-35mA, then we are looking at maybe 1 milliwatt of dissipation, correct?
P = I^2 x R = (0.03)^2 x 1 = 0.0009 watts.

Any reason I couldn't use a 1/2 watt 1% tolerance metal film?

That is supposed to be 1/2 watt 1 ohm

I'll go change it

[pdf64]

You don't need a diode - if a tube shorts and the 1 ohm resistor opens  that is a good thing.

I don't see it like that, better for a HT fuse to blow.
Or failing that, maybe a marginally rated screen grid resistor (if that's the fault path).
For amps lacking a HT fuse ( ), a significant downside of the cathode resistor blowing is that the fault current then stresses the next weakest link in its path to 0V common, which is often the cathode-heater insulation.
Depending on how the heater circuit gets its reference, when that happens, it can cause further collaterol damage or just stress everything in its path. Which includes the expensive stuff.

[sluckey]

By the time that diode becomes forward biased the tube will be melted down due to insane cathode current.

[PRR]

Historically low-Ohm resistors were not made in small sizes. As you see, this job can't need a whole Watt.

Using a diode to "protect" a resistor seems like false thinking to me. It may weasel some flame test: diodes and even tubes do not fail in flames, while older/cheaper resistors did.

As pdf says, letting cathode float high may ultimately burn a power transformer which IS stinky and panicy. This be a good reason to over-oversize the resistor.

[PharmRock]

Hey guys thanks for all the info.  Based on the size of the resistor, I believe I have a 1-watt, which is what's on the board now.  I think these were pulled from the pre-populated Hoffman Plexi-100 board when I converted it to a 50-watter. 

[pdf64]

Big resistors can roast stuff pretty good https://www.thegearpage.net/board/index.php?threads/6g3-brown-deluxe-build-done-final-pix.2305389/#post-33586721