Resistor voltage rating

[Smokebreak]

When choosing max voltage rating for say, PS resistors, are we concerned with voltage before or after the resistor, or the voltage drop across the resistor?
I thought it was voltage drop, but I see in the BOMs that the droppers called for are the 750V variety, so I'm just curious.

Also, do you guys like MF or MOX in the PS. I noticed BOMs state MO, but Mouser part # takes you to Vishay MF??

Thanks,
J

[VacuumVoodoo]

In 95% of cases operating voltage across resistor is the limiting value. You also need to take power dissipation into account since P=U²/R or U=SQRT(P*R)
In some cases initial voltage applied to the resistor at turn-on can be higher than its rating, then you need to check the resistor's datasheet for max isolation voltage.

[Smokebreak]

I see, thanks.

You also need to take power dissipation into account since P=U²/R or U=SQRT(P*R)

So while I'm dropping 85V over 18K, 85*85/18000= ~.4W
I say this because I've got a bunch of 2W in a PS and was thinking of changing to 3W as most suggest, but 2W should be fine for the long haul, eh?

In some cases initial voltage applied to the resistor at turn-on can be higher than its rating,.

But how much are we possibly talking about here, in terms of figuring wattage? I understand the max voltage and isolation voltage, but if we're figuring for more than a 350V rated resistor, say, then shouldn't we also plan on at least that for power handling?
350*350/18000 suggests a 7 watt minimum resistor in that position, and in this case I'm talking about inbetween screen and PI node, and usually that calls for 3W(or hopefully 2;)).
This is where I'm a bit confused.

[tubeswell]

Say you have a 5F6A with a 5AR4 rectifier. Say that when you switch it on, you do so with the standby straight away (i.e.; a 'cold start'), the B+ comes up very slowly as the rectifier heater warms up. In this case there is no 'start-up' surge, so the voltage drop across the supply resistors never gets above what it would be in the amp's normal operating mode (which is anything between about 10V to about 50V or so, depending on the resistance of each respective supply resistor and its sequential position in the supply chain).

At the other end of the rectifier scale, in an amp with SS diode rectifiers, there is a startup surge where the B+ rises very fast to the maximum possible with the PT's high tension winding's secondary voltage, and then settles down as the heaters warm up and brings the voltages in the B+ rail down to their respective operating voltages (which is anything between about 10V to about 50V or so, depending on the resistance of each respective supply resistor and its sequential position in the supply chain). In this scenario there is no voltage drop across the supply resistors at startup, because there is no B+ current at first (because the heaters are still cold), so the voltages on all sides of the supply resistors are all the same.

So why the concern about maximum voltage rating?

[Structo]

Yes it depends on the amp design and type of rectifier used.

In power supplies I usually use 2W or 3W metal oxide (flameproof) resistors.
Or even the sand box wire resistors, to dissipate the necessary heat.

I believe in making the power supply very robust for reliability and fewer trips to the bench.
It really doesn't cost all that much more unless you are building hundreds of amps.

[Smokebreak]

Say you have a 5F6A with a 5AR4 rectifier. Say that when you switch it on, you do so with the standby straight away (i.e.; a 'cold start'), the B+ comes up very slowly as the rectifier heater warms up. In this case there is no 'start-up' surge, so the voltage drop across the supply resistors never gets above what it would be in the amp's normal operating mode (which is anything between about 10V to about 50V or so, depending on the resistance of each respective supply resistor and its sequential position in the supply chain).

At the other end of the rectifier scale, in an amp with SS diode rectifiers, there is a startup surge where the B+ rises very fast to the maximum possible with the PT's high tension winding's secondary voltage, and then settles down as the heaters warm up and brings the voltages in the B+ rail down to their respective operating voltages (which is anything between about 10V to about 50V or so, depending on the resistance of each respective supply resistor and its sequential position in the supply chain). In this scenario there is no voltage drop across the supply resistors at startup, because there is no B+ current at first (because the heaters are still cold), so the voltages on all sides of the supply resistors are all the same.

So why the concern about maximum voltage rating?

Ok so am I understanding correctly that it is with SS rec that I should be using 750V,say?
I just had a slight freakout moment doing a Mouser order and examining the datasheets a little closer, and then I realized I was unclear on the subject.

I believe in making the power supply very robust for reliability and fewer trips to the bench.

and this is great point I used a 10W 1K on my 6V6 Express, and had no problems with it...till it burned up after 3 weeks. I actually laughed out loud when the 25W got here and I had to try and figure how to squeeze that in

Thanks guys

[Smokebreak]

Ok so looking at this resistor : http://www.mouser.com/ProductDetail/KOA ... tKT%2fI%3d
and it's datasheet: (page 2, item MOS3): http://www.mouser.com/ds/2/219/MOS-16613.pdf

, 350V is the most that can be dropped accross it, and 700V would be the max for a surge?

[tubeswell]

Ok so am I understanding correctly that it is with SS rec that I should be using 750V,say?

No you typically wont see more than about 50V drop on any supply resistor, even with a SS rectifier. The voltage drop on startup is nil, because there is no current in the B supply until the heaters warm up. So the entire HT supply (including all the supply resistor nodes) is sitting at the same voltage. So there can't be any voltage drop.

The only place you could see a high voltage drop is where you might have bleeder resistors between the B+ and the ground return. But otherwise not in the supply resistor chain

[Smokebreak]

Ok I've got it now. Thanks a bunch.

[martin manning]

No you typically wont see more than about 50V drop on any supply resistor, even with a SS rectifier. The voltage drop on startup is nil, because there is no current in the B supply until the heaters warm up. So the entire HT supply (including all the supply resistor nodes) is sitting at the same voltage. So there can't be any voltage drop.

The only place you could see a high voltage drop is where you might have bleeder resistors between the B+ and the ground return. But otherwise not in the supply resistor chain

Start-up with a SS rectifier is the worst case for voltage drop across these resistors because a filter cap initially looks like a dead short to ground. The first resistor in the string is going to see the full B+ voltage dropped across it momentarily.

[katopan]

I learnt this from blowing trasistors in a SS amp (battery guitar) I built years ago, but all the resistors do see voltage on power up before the valve start conducting current. Before the coupling caps are charged they have zero volts across them. Current flows through them to charge up and in the mean time the B+ is across the complex resistive network of power rail resistors, plate resistors and plate load resistors & pots (after coupling caps to ground). These charge up pretty quick so they don't see the voltage across them for long, but it does happen on power up. Obviously a valve recto softens this significantly because it allows the voltage to come up slower (drops supply before/as it warms up) and limits the current into charging up the supply and coupling caps. Also got to account for charge up time of the supply filter caps, but often this is quicker than the warm up time of the valves, so you still get charge up of the coupling caps before full valve conduction.

[tubeswell]

Start-up with a SS rectifier is the worst case for voltage drop across these resistors because a filter cap initially looks like a dead short to ground. .

Ah yes sorry. I forgot about the filter cap sitting at zero voltage

[Smokebreak]

Ok now that we're in agreement, how are these for PS :

Ok so looking at this resistor : http://www.mouser.com/ProductDetail/KOA ... tKT%2fI%3d
and it's datasheet: (page 2, item MOS3): http://www.mouser.com/ds/2/219/MOS-16613.pdf

, 350V is the most that can be dropped accross it, and 700V would be the max for a surge?

I'm just glad I can barely comprehend these discussions these days
I think this sums it up nicely from the AX84 FAQ :

Voltage rating : as with wire insulation, make sure this will handle the voltages you care about. In most resistors, you only care about the voltage drop across the resistor, but if the resistor is up against, or can possibly touch, another conductor (such as a ground lug!) or another resistor, you have to worry about the maximum potential between the resistor and the other conductor or resistor.

[Structo]

Yes that is why I said 2W-3W for the PS resistors.

I believe that a 700v resistor is most likely safe in the power supply unless you are running 500+V on the plates.