Running 4 ohm speakers with my Simple SE

[Coreyk78]

Hello all,
I have a Simple SE that I built a few months back and I have been very happy with.

Currently I am using some Edcor GXSE 15-8-5k OPT in triode mode and they seem to work great running the 8 ohm and 6 ohm speakers that I have.

However, I've started gathering parts to build a pair of Zaph BAMTM speakers which are a solid 4 ohm speaker, so I'm thinking of changing my OPTs for something that can better handle a 4 ohm load. Here are my questions:

1. What would happen if I tried running 4 ohm speakers with what I have now?

2. If I switch to transformers that have a 4 ohm secondary, like some big Edcor CXSE 25-4-5k will I have weak power output into 8 ohm speakers?

3. I know the Hammond 1628SEA has multiple taps for different speaker impedance, which would be nice for me, but they are kind of expensive for my budget. Anyone know of a cheaper option?

Thanks for any help, its probably obvious by my questions that I don't really understand how output xformers work, I've just been following other's recommendations on here, hehe.

[rman]

Hi.

Edcore does custom power transformers at a reasonable cost. Maybe the would make you an output with a 4 ohm tap as well?

Rolf.

[boywonder]

Hooking up 4 ohm speakers to your 8 ohm taps will half the primary impedance of your transformer from 5K to 2.5K. If you look at the "tubes and transformers" page on Tubelab George's website, the tables show simulations with 5K, 3K loads for the tubes. Notice at high B+ and 3K loads, most of the tubes are exceeding their dissipation limits as shown by the red font.

So your 2.5K load is a little less than a 3K load.........

So running 4 ohm speakers on your 8 ohm taps is certainly do-able with a low enough B+ voltage, resulting in a little more power out and a little more distortion, depending on the B+, tube, and load impedance.

For question #2, yes, the power into 8 ohm speakers on the 4 ohm taps would be reduced, since the transformer primary would effectively be 10K (5K x 2)

[Coreyk78]

Well following the chart, it looks like I can get away with using my current transformers. I have a pair of chinese 6L6GCs right now, and 620 ohm bias resistors so it looks like it should be borderline ok with a GZ34 rectifier, otherwise I have some other rectifier tubes to drop the B+. And I have been looking at trying some KT88s and it would be more than safe with those tubes.

Thanks for pointing me in the right direction boywonder!

[Ty_Bower]

Speaker impedance curves are anything but flat, and vacuum tubes are very forgiving with respect to specification tolerances. In other words, it'll probably be just fine and in all likelihood you won't even be able to hear the difference.

The transformer is mostly just a ratio device. With regards to impedance ratio, a 5K:8 transformer is the same as a 2.5K:4. The ratio of the number of turns on the primary winding to the number of turns on the secondary winding will be the same. Look at the datasheets, examine the plate curves. Figure out where your loadline will be at the equivalent load and decide if you are happy with it. The general guideline is that higher loads give less distortion, and lower loads can extract more power.

I did say the transformer was mostly a ratio device. There's also an inductance component (good thing) and a DC resistance component (necessary evil). To get more inductance, you need more turns (and possibly a bigger core). But, as you add turns you increase the DC resistance. More inductance gives better low frequency response. Higher DC resistance robs power and reduces efficiency. There's also things I don't understand, like leakage inductance and shunt capacitance.

At a relatively high impedance, you need less primary inductance to get a desired low frequency response. I think the formula is Fo = Z / (2 * pi * L). If you use an output transformer with sufficient inductance to achieve the desired low frequency response at a high impedance load, it should have more than enough inductance at a lower load. In other words, the low frequency response will be even better. The cost is additional dB of insertion loss from the many turns in the winding. Yet another way of saying things is that output transformers "ratio down" relatively well, maintaining (nay, improving!) their low frequency response in exchange for a small loss in power. On the other hand, they don't "ratio up" as nicely, potentially losing precious Hz of response on the bottom end due to insufficient inductance.