OT questions

[rf7]

Does the primary impedance of the OT affect the output power of the amp?

If so, are they inversely proportional? (i.e., Higher primary impedance means lower output power)

Should OT's be over rated for the output power? Or do the OT designers already do that?

[Enzo]

An OT, or any transformer for that matter, is like a set of gears. it converts or transforms energy. In this case it transforms the high voltage, low current signal at the power tubes into the low voltage, high current signal the speakers require.

What is important is that it transforms in the process the impedance of the speaker to the impedance the tubes want to see. Transformers have no inherent impedance, they only have turns ratios. We may say this transformer has a primary impedance of 5000 ohms and a secondary impedance of 8 ohms, but the ONLY time the primary sees 5000 ohms is when there is an 8 ohm speaker across the secondary. Put a 4 ohm or 16 ohm speaker there, and the primary impedance changes too.

So the primary impedance can affect the power output insofar as it matches or does not match the speaker impedance at its secondary. Use the wrong transformer ratios and the system won;t transform the tube circuit power as efficiently. So it is not about higher or lower, it is about matching.

Think of the transmission in your car. Your engine has X amount of horsepower, but using 5th gear to start out does not transform the engine torque into much useful power at the wheels. Likewise 1st gear is not the best way to get highway speeds. But neither gear affects horsepower, just how well it gets to the wheels.

[tubeswell]

Should OT's be over rated for the output power? Or do the OT designers already do that?

The more iron and thicker the wire in the windings, the more power the OT can transfer efficiently. Big iron reproduces better bass frequencies because the core has more mass to handle the induction of the current - because there is more current hanging around during the longer (bassier) signal wavelengths. Whereas smaller iron would saturate sooner - reducing fidelity (which is not necessarily a 'bad' thing as far as guitar amps go). Well that's how I see it anyhow.

So it depends on what you want the OT to deliver - cleaner lows etc, or more 'dirt'.

[Steve Conner]

If so, are they inversely proportional? (i.e., Higher primary impedance means lower output power)

As Enzo hints with his gears analogy, it's not quite as simple as that. For a given tube type and B+ voltage, there'll be an optimum load impedance that wrings maximum power from the tubes. And it can vary a lot: for instance if you wanted to use a pair of 4CX250B tubes, usually used by radio hams, then the maximum power you can get is around 600 watts. And to get it you need over 2000V B+ and a load impedance of nearly 10k.

But yes, the general trend in guitar amps is that higher powered ones have a lower primary impedance. This is because they all tend to use about the same B+, and the higher power ratings are achieved by adding more parallel pairs of tubes.

Should OT's be over rated for the output power? Or do the OT designers already do that?

Again it depends what you mean by overrated, and who the designer was. A hi-fi style OT like Hammond's standard range will use a generously-sized core to make lots of undistorted bass. Hammond used to say that you could use these same OTs for guitar amps, and just double the power rating, so their 50 watt model would suit a 100 watt guitar amp. But now they make dedicated guitar ones that are supposed to have more mojo, or something.

On the other hand, if you look in an old Silvertone, the OT is a measly little thing, and probably well underrated for the job. But some players like these amps.

[Redhat]

Also, keep in mind that tube amps act like current sources while most SS amps act as voltage sources.

A perfect voltage source provides the same voltage regardless of load resistance. Real V sources (like batteries) love open circuits or those with high impedance since they don't have to create as much current. Real V sources hate short circuits or those with low impedance since they must crank up their power so that the voltage remains constant.

On the other hand, a perfect current source provides the same current regardless of load resistance. Real I sources hate open circuits or those with high impedance since they have to crank up their power (by way of output voltage) to provide a constant current value.

This is why its important to match the output impedance of the tube amplifier with that of its load. The tube wants to produce a relatively constant current from its anode over the plate's voltage range as the OT's primary's impedance changes with frequency - like a current source. Providing too high of an impedance (or disconnecting the speaker leaving the secondary open) makes it harder for the tube to act like a current source.

And unhappy tubes don't live long.

[clyde1]

Is there a correlation between plate voltage/plate current and optimum impedance of an output transformer for a given tube? Rules of thumb for different tube types, ie pentode, beam-power?

[ThermionicScott]

Is there a correlation between plate voltage/plate current and optimum impedance of an output transformer for a given tube? Rules of thumb for different tube types, ie pentode, beam-power?

It depends whether you intend to run the plate(s) and screen(s) at the same voltage or not (for example, most BF/SF Fenders run the plates and screens at roughly the same voltage, but the older wide-panel tweeds run the screens much lower than the plates). To get the most power out of an output tube, you want the load line to intersect the curve at the "knee".

If you stick to the voltages in the tube manual, you can use the loads they recommend. If you want to do your own thing, you might be cheating yourself out of some watts if you get stuck on rules like "6V6's need an 8k load".

What kind of output section do you have in mind?

- Scott

[Chuck H]

Something that has been implied (tubes are a current source, therefore...) but not specifically mentioned is that in the right conditions a lower primary impedance will increase output power AT THE EXPENSE OF LINEARITY. Linearity is not as much of an issue with guitar amps as with hi fi amps. I think THIS is why some guitar amp MFG's have used lower than tube data specs for the OT primary impedance.

Chuck

[rf7]

Thanks to everyone for the replies.

[Steve Conner]

Is there a correlation between plate voltage/plate current and optimum impedance of an output transformer for a given tube? Rules of thumb for different tube types, ie pentode, beam-power?

Yes. If you know the maximum peak signal current that the tube is rated for, then the optimum impedance is about 4 times (your B+ voltage)/(peak signal current) or usually a little less, because the tube still drops some voltage when it's turned on as hard as it'll go.

So for instance, the EL34 datasheet gives a max signal current for a push-pull pair of 300mA. That's an average as you'd measure with a meter when the tubes are amplifying their maximum unclipped sine wave, so the actual peak current is 1.6 times that: 480mA, or 0.48 of an amp.

If we say our B+ is 450V, then the optimum impedance is then roughly 4*(450/0.48) = 3750 ohms. Which is about what most EL34 amps use, give or take that "little less".

The factor of 4 is because OT impedances are quoted from end to end of the primary, but the calculation gives us the impedance from the centre tap to one end, and the a-a impedance is 4 times that.

[trem]

Also, keep in mind that tube amps act like current sources while most SS amps act as voltage sources.

A perfect voltage source provides the same voltage regardless of load resistance. Real V sources (like batteries) love open circuits or those with high impedance since they don't have to create as much current. Real V sources hate short circuits or those with low impedance since they must crank up their power so that the voltage remains constant.

On the other hand, a perfect current source provides the same current regardless of load resistance. Real I sources hate open circuits or those with high impedance since they have to crank up their power (by way of output voltage) to provide a constant current value.

This is why its important to match the output impedance of the tube amplifier with that of its load. The tube wants to produce a relatively constant current from its anode over the plate's voltage range as the OT's primary's impedance changes with frequency - like a current source. Providing too high of an impedance (or disconnecting the speaker leaving the secondary open) makes it harder for the tube to act like a current source.

And unhappy tubes don't live long.

This is interesting stuff. So let me ask then, in theory, assuming you have a 4k primary with an 8 ohm tap.
1. Is it "worse" if the secondary sees 32 ohms than if it sees 2 ohms..?
2. Assuming you have the above situation, what will fail first, the OT or the power tubes..?
Thank You

[Steve Conner]

Like Redhat hinted earlier, solid-state amps and tube amps are duals of each other.

The solid-state amp is a voltage source, and you destroy it by loading it with too LOW an impedance. (Though better quality solid-state amps will automatically limit or shut down to protect themselves.)

The tube amp is a current source, and it hates too HIGH an impedance. (But it doesn't much like one that's too low either.)

The usual failure modes that I know of are:
When a tube amp gets cranked up into too high an impedance, the tubes' screen grids draw very heavy current, melt, short against some other electrode and blow the screen resistors.

Or the tube sockets or OT arc over internally and burn up.

If you load a tube amp with too low an impedance, the tubes will overheat and wear out prematurely, but it's not as damaging as the too-high case. If you want, you can hook a 2 ohm cabinet to your 8 ohm tap, crank it, enjoy the weak, crappy sound and watch the plates get nice and red. So, 32 is worse than 2.

[trem]

Like Redhat hinted earlier, solid-state amps and tube amps are duals of each other.

The solid-state amp is a voltage source, and you destroy it by loading it with too LOW an impedance. (Though better quality solid-state amps will automatically limit or shut down to protect themselves.)

The tube amp is a current source, and it hates too HIGH an impedance. (But it doesn't much like one that's too low either.)

The usual failure modes that I know of are:
When a tube amp gets cranked up into too high an impedance, the tubes' screen grids draw very heavy current, melt, short against some other electrode and blow the screen resistors.

Or the tube sockets or OT arc over internally and burn up.

If you load a tube amp with too low an impedance, the tubes will overheat and wear out prematurely, but it's not as damaging as the too-high case. If you want, you can hook a 2 ohm cabinet to your 8 ohm tap, crank it, enjoy the weak, crappy sound and watch the plates get nice and red. So, 32 is worse than 2.

Thank You -
If I can ask another question....why is it then that we always read about the danger of burning down Output Transformers (especially with no load)..? I know I am not the only one who has read a query by somebody on a forum that was worried he damaged his OT because "At a gig last night the bass player tripped over my speaker cable and yanked it out of the head. I played like that for 30 seconds". And not just that scenario. It seems whenever somebody has a impedance question regarding their speakers and OT, it is always the transformer that is said to be in danger..?
Again, I am not making a statement, I am just asking. With the situations that have been discussed in this thread, is it the tubes that will likely fail before the transformer? Do OT's get ruined very often by what would be a likely impedance mismatch caused by guitar players.?
Thanks

[Bruce / Mission Amps]

Thank You -...I am just asking. With the situations that have been discussed in this thread, is it the tubes that will likely fail before the transformer? Do OT's get ruined very often by what would be a likely impedance mismatch caused by guitar players.?
Thanks

My gut and personal experience says no, the OTs are not ruined very often at all by speaker or load mismatches.
I think any load is so much better then no load that those statements might be a myth.

BUT... unlike RG Keen and a few others here, I am not a "magnetics" engineer so I can't give you the hard core science on it.
Regardless, I think a problem does exist where some transformers are made with fairly thin MIL spec wire and cheap insulation (which over time has degraded and become defective) and a player uses an impedance load that is much to high causing a higher AC voltage to develop across primary windings.... which can flash over and subsequently, shorts the primary windings to each other.
That's bad because one side of the primary (with respect to it's center tap) has a different turns ratio now and now a different impedance ratio too.
It can sound bad and could eventually lead to one side blowing open.
Also, the flash over can create a high voltage spike through the OT that will cause a corona discharge or plasma beam jolt on the socket too.
Such that it will jump over the plate lug to a low impedance lug/spot on the socket, such as a very close by filament lug.
Then it also can ruin the socket leaving carbon scoring which continues to cause flash overs... ruining the tube or causing it to draw too much current.
I'm sure there are other failure modes that can be found too.

[Enzo]

why is it then that we always read about the danger of burning down Output Transformers (especially with no load)..?

Because NO load is not an issue of "wrong" impedance. With no load, there is nothing to stop the inductance of the transformer winding from kicking up voltage, and you can wind up with voltages way over the capacity of the winding insulation to handle. The transformer then arcs internally and is destroyed.

That is a totally different ballgame from mismatched impedances. Most cabs will be of the 4/8/16 ohm type. I am sure you can find a 2 ohm cab or a 32 ohm cab, and maybe a little less rare might be something like a 5 ohm cab. Well 5 point something. If you connect a cab that is off one step either way on that chart, it is not very likely to hurt anything. Might not work its best, but it won;t hurt it. get too far off, and it threatens the power tubes for reasons discussed above by others, but not really the transformer. In my opinion it is the open circuit - a lack of load - that threatens the transformer itself.

Transformers are almost always the lowest thing on my list when i enco0unter amp problems. Tubes are almost always the first thing on the list of failures.