Yet Another Speaker Shootout - G12 65 versus WGS ET65

[BobW]

Maximum power theoretically occurs when you match the secondary impedance to the load. 16 Ohm spkr to the 16 Ohm tap, 8 Ohm spkr to the 8 Ohm tap. However both the primary and secondary impedances are constantly changing due to reflected impedances from both the tube and speaker.
Since most of the Lounge article dealt with UL taps, it doesn't apply to Suppressor Grids when connected to a DC power tap. The impedance change here is less than with a UL transformer. The purpose of the UL taps are to self regulate and optimize the impedance to provide maximum efficiency. This is the reason why UL designs can provide more power and greater efficiency than a non-UL design.
I see no reason why you can't use 2 different speaker impedances (16 and 8 ) to the 16 and 8 Ohm outputs. If someone else can offer a real valid reason why not, then please prove me wrong. A simple test tone and scope setup would prove if the output power is being distributed equally, but the real test here is how does it sound?

[talbany]

However both the primary and secondary impedances are constantly changing due to reflected impedances from both the tube and speaker.

Verifying Bob's analogy a bit further
Take an average beam power output tube, such as the 6V6 requires a load of around 5,000 ohms, Plug in an 8 ohm speaker into an 8 ohm tap now send a 1kHz signal through it.. Perfectly or well matched..at 8 ohms..
The issue is guitar's frequency ranges from approx..82 to 1050

Everyone knows that the impedance of a loudspeaker is frequency dependent, meaning that at each frequency the impedance has a different value.(Higher frequency's higher impedance) Therefore loudspeaker manufacturers give a mean impedance value at the same 1kHz frequency.. The consequence of this is that you never can calculate exactly the value of the primary
impedance due to a wide range of frequency's generated, Just try to be close to the primary and secondary impedances intended in the design, but don’t worry for deviations up to about 20 %....Not to bring up an old topic..Output transformer impedances is NOT an exact science just an approximation...Deal with it..LOL!!

Tony VVT

[txbluesboy]

Lets take a transformer with 4-8-16 ohm secondarys and a primary impedence of 4000 ohms. When you plug an 8 ohm speaker into the 8 ohm tap you are reflecting 4000 ohms back to the power tubes. the same goes when you plug a 16 ohm speaker into the 16 ohm tap. However when you plug into both taps your primary impedence drops to about 2000 ohms. This is why I think you should plug the speakers into the lower taps. On this same transformer when you plug a 16 ohm speaker into the 8 ohm tap you will reflect a primary impedence of 8000 ohms,right? Then when you plug the 8 ohm speaker into the 4 0hm tap you will bring the primary impedence back to 4000 ohms. As stated earlier tube output transformers are not an exact science. 100% mismatch is normally considered ok. You would have to try it both ways to see which sounds better.

[BobW]

...On this same transformer when you plug a 16 ohm speaker into the 8 ohm tap you will reflect a primary impedence of 8000 ohms,right? Then when you plug the 8 ohm speaker into the 4 0hm tap you will bring the primary impedence back to 4000 ohms...

Not really, since reflected impedance is not a linear function. As Tony stated that would only be vaild at a particular frequency, say, 1kHz.

The general consensus is that a 100% mismatch won't damage your speaker or OT, so trying a combination of speaker impedances with their respective taps and deciding what sounds best is a personal choice.