Of fuses and current draw

[xtian]

I have a Mesa Heartbreaker 100watt head that blows fuses if you're not careful to employ the standby switch at startup.

It uses a 3 amp slo-blow fuse. Why? When EL34s are rated for 1.5 amps each for the heaters alone? And it also has a 5AR4 rectifier AND SIX preamp tubes!

But then, my JCM 800 2203 (100 watt) uses only a 4amp fuse.

How is this possible? Calculated current draw of almost 8 amps, but only a 3 amp fuse? Is that because it's lower amperage on the PRIMARY side at 120vac, but higher amperage on the secondary side?

[Leo_Gnardo]

The AC line current has more to do with the sum of all the power, current times voltage, drawn on the secondary. And you have to add that up for each winding. Lotta arithmetic. And ... that calculation is valid only at steady-state operation, assuming the filaments are already heated up. When they're cold, on startup they draw a lot more current if only for a fraction of a second. I've measured startup current pulses up to 6X the steady-state figure, a good reason to have slo blo fuse in the AC line.

Not clear if your fuse-popping is on startup or when you go from standby to run. If it's on startup, the fuse rating really is marginal if charging the hi voltage supply is enough to pop the fuse, but doesn't if it's disconnected with standby mode employed. All in all, I'd go to 4A and fuggeddaboudit.

100W Marshalls, as you noted, generally carry 4A mains fuses and that's my logical shortcut once all is said & done.

[xtian]

Yes, fuse pops on startup. Thanks for the confirmaion; I was thinking 4a fuse, too.

[tubeswell]

Sum the (maximum) VA of each of the secondaries, and then divide this by the voltage inn the primary to get the current draw on the primary. Rate the mains fuse at 2 x what that primary current draw is.

[pdf64]

How is this possible? Calculated current draw of almost 8 amps, but only a 3 amp fuse? Is that because it's lower amperage on the PRIMARY side at 120vac, but higher amperage on the secondary side?

Yes, the power transformer, err, transforms, such that
8A @6.3V = 50W
50W @ 120V = 0.42A

Added to that ac line current will be losses in the transformer, eg 10%, so maybe 0.46A.

So that's constant load drawn by the heaters from the line.

My view is that the smallest value fuse that will support cranked operation is the best option; think how sick you will feel if, after fitting a 4A fuse, a tube short causes a prolonged fault current that damages a transformer winding.
If a high quality slow blow eg Bussman or similar, blows at start up, consider additional in-rush current mitigation such as NTC thermistors, eg http://uk.rs-online.com/web/p/thermistors/5167827/
They can be fitted in series with the primary circuit and / or the secondary eg B+.
Pete

[teemuk]

Calculated current draw of almost 8 amps, but only a 3 amp fuse? Is that because it's lower amperage on the PRIMARY side at 120vac, but higher amperage on the secondary side?

Do the math.

Theoretically you can start from assumption that power (UxI) at both primary and secondary side is the same.

Since result of P=UxI is the same for both primary and secondary BUT the voltage (U) on both secondary and primary varies it means the current must vary as well. A good rule of thumb = The lower the voltage, the higher the current draw. Basic math.

You must calculate the power draw of all secondary circuits and then use the sum of power to estimate power drawn from the mains. This will only land you to ballpark constant power draw value. It will not be overly helpful...

...Because in practice you will also encounter issues like losses and surge currents. Losses are simple: you never get 100% power use so to produce x amount of watts at secondary the primary draws x times efficiency amount of power from the grid. Surge currents are more difficult to deal with: For example, a typical power supply rectifier/filter circuit never draws constant current, it draws momentary current surges and these are usually much higher in amplitude than your calculated current draw, which is more of an average, nominal value. Think of arriving to that value from the average value of transient currents levelled over certain time period. It's not constant DC draw so the basic math will only get you this far.

You will likely need to apply some waveform arithmetics to find out to how high surge currents those nominal current draw values convert. The problem is, the calculations are very much amp specific since you need to consider variables like VA rating of the PT, surge current characteristics of the power supply, filtering capacitance, etc.

[Reeltarded]

Or you could just add .5A to the fuse instead of understanding all that.



S'what I do.

[teemuk]

^ Or a little metal bar.

Who cares that fuses actually try to protect from things like electrical fires or annoying faults like dead $$$ power transformers.

[John_P_WI]

It's the damn near infinite current draw of the big iron inductive load that's killing the fuses. Most of the above is talking about steady state conditions and calculations.

I have seen resistors placed in the primary side used to limit the initial inrush current to the PT. There is a reason the lights dim when you first flip the switch...

[Reeltarded]

I didn't say go up 10A!

There is a differwnce in making a small move up and shorting it with a screw. If the amp normally runs but spits out a fuse every other time the switch is flipped I throw another half watt at it. I wouldn't do this if I suspected a problem other than the fuse, like hot caps or the transformer is hot enough to toast bread.

[mhartman]

Or you could just add .5A to the fuse instead of understanding all that.



S'what I do.

+1 On a custom build, once the amp is confirmed not to have a short (light bulb tester), if I have a problem with the fuse (I start low), I typically up it a bit at a time until I get it right. That being said, I'm a hack who typically avoids letting too much math into my shop area.

[Reeltarded]

+1. I am also a mathless hack.

[xtian]

I appreciate both the maths and the MacGyvers in da hizzy. I've got a gum wrapper around here somewheres...

[Leo_Gnardo]

I've got a gum wrapper around here somewheres...

Gum wrapper hell. Snip & strip 1.25 inches of 12 ga Romex, then give the amp to somebody you really don't like at all.

Some years back a customer gave me a troublesome amp to fix, should have had a 3A fuse. His uncle "the experienced electrician" tried to fix it. Oh he fixed it all right. With a 30 amp fuse! Well at least it wasn't popping fuses anymore. Don't mind the stinky smoke coming out of the PT... or the loud hummmmm. Or having to reset the breaker each time you fire the amp up, or the smoke coming out of the wall.

------

By now you've got the hang of both methods. A transformer is sort of an electric "gearbox." Power in = power out, less a tiny bit for transformer ineffiency. Multiply amps calculated by fudge factor for the startup current pulse.

Or take a cue from commercially made amps of similar power.

And you can always hedge your bet like old Ampegs. Put an extra fuse inline, inside the amp, rated a bit higher than the fuse in the panel holder.

Then, for transistor amps, there's the old definition of a transistor: a device that blows up to protect fuses in the circuit.

[teemuk]

Or take a cue from commercially made amps of similar power.

It only works if total power drawn from mains by both amps is about the same, the specs of the OT and power supply in both amps about the same, and possible schemes to limit inrush about the same.

If these happen to be different (which they likely are) then there's very little point in trying to copy a fuse value from another design of similar power.

For one...

- Output power rating does not indicate how much power the amp actually draws from mains. A 100W solid-state class-D amp may draw a little over 100W from the grid while a 10W class-A tube amp may easily draw as much. You need to consider amplifier's efficiency as well as things like total current consumption of circuit elements like all circuits, indicator lamps, filaments etc.

- Output power rating does not indicate anything about transformer specs. Some transformers have way higher inrush than some others (e.g. toroidal vs. EI core, low VA rating vs. high VA rating) and some amps may have built-in inrush limiting whereas some other amps may not. Copying a fuse value from another amp with similar output power may be entirely pointless if the other amp happens to have higher or lower inrush than the other.

- Output power rating doesn't tell anything about power supply specs like overall capacitance. This is once again one of those specs that has huge effect on selecting a proper type and properly rated fuse.