WT-550 Back Panel

[CoolBlueGlow]

Reasonable conclusions IMO, though a toroidal transformer isn’t necessarily more or less efficient because it’s a toroid. It depends on the overall design, the core material and how much cooler is used.

Also, if you are a commercial service tech, you should update your forum profile disclosure to reflect being a commercial member under the CUP.

Yup, agreed. As I said... "can be", not "are". In this case it looked to me like maybe they are better? Guessing Eden did some work to make WT550 torroidal version better, maybe because they came to the same conclusion on the bench as anyone would about the E/I core version of the PSU on the WT550? (a guess here, of course. It is possible that E/I core transformers got more expensive or unavailable or whatever.)

I can imagine such a conversation at Eden back in the day. "Today's issue for product improvement; WT550 amp runs out of power supply ooommph at something less than it might with a better power supply."

I could be wrong, of course, but something is better about those later designs. That, or my equipment is miscalibrated or the equipment of the fellow on YouTube who performed a similar experiement (or both). He got 376w with a dead load. I got 306w. Same preamp, same amp board, but different power supply. Dont' THINK my stuff is that badly miscalibrated. It would require three different devices all simultaneously wrong in exaclty the same error direction, and by more than 20%. Doesn't seem likely, but not impossible.

Based on basic confidence in my methods and equipment, and the observeation of another similarly competent individual's experimental results, I surmised the missing power had to come from somewhere, so I made an educated supposition on the power supply. I freely admit is is Informed conjecture, but in my defense, it is reasonably based on fifty plus years of training and experience.

But I must be careful, as Enrico Fermi said, "The first principle is that you must not fool yourself, and you are the easiest person to fool."

Three possibilities; 1.) the power was mismeasured by one or both parties, or 2.) The extra power is there in version B and not in version A, or 3.) "other".

re: "commercial service tech". Well, these days I am in the enviable position of not having to work on anything I don't want to work on. It is a blessing, but thanks for suggesting that I revise my profile disclosure.

I guess I can put "retired solder sniffer" or something harmless like that.

 

[agedhorse]

To be off by 20% in power doesn’t mean that you are off by 20% in measured voltage because of the squared voltage in the power equation.

The measured voltage for 300 watts 4 ohms) would be 35V and for 375 watts would be 38.5V… only a 3.5 V difference.

When measuring power of any power amp that uses an unregulated power supply, it’s critical to monitor the input power’s wave form for flat topping of the voltage waveform. This is very common as current increases due to I^^2 x R losses in the primary circuit as well as the effect of the non-linear load the a FWB imposed on the primary load. The peak charging currents cause voltage drop of the peak voltage of the waveform, and the audio power delivered to the load is directly impacted by the PEAK voltage of the input power waveform.

 

[CoolBlueGlow]

Yeah, I reckon so. But let me be clear that I meant measurements which by calculation suggested ~20% less useful output power, not 20% less measured voltage. That is what I deduced by my methodology. I checked useful output power in (at least) two ways. One was the voltage measurement, one was power supply efficiency vs. maximum consumed current. One method was direct, one was indirect. Both were then extrapolated mathematically and results were compared. All agreed in general about what I deduced to be 20% less useful power in E/I vs Torroid transformer equipped design (sample size 1)

In either case, that is significantly less than the 550W output which Eden marketing suggests. Of course there could still be something wrong with the one amp I measured, but it does seem to be working and now gigging again with no complaints from the owner.

Frankly, I feel sorry for Amp manufacturing engineers. The whole how many watts game is silly and specious. Engineers work hard to make an amp SUFFICIENT and MUSICAL, like this Eden WT-550 seems to be. Then some goofball in a music store reads a one page marketing sheet and says "Gee, Amp A is better than Amp B because A has more watts".

But I did all that roundabout stuff specifically because the other guy's torroidal transformer equipped amp was clearly making 3-ish more volts than my E/I equipped unit, just as you predicted, when tested just as he tested his amp. Virtually the same measurement setup - I got 35 volts, he got 38ish.

The power supply efficiency calculation also pointed at around 300-ish watts max output @ something less than 100% duty cycle, which I took to be a kind of alternate confirmation of my first measurement of 35V

From these two observations I make a general term of conversation, i.e. "20% less power in the older E/I version of these two otherwise similar designs".

In the real world, with impulse signals like a real bass player generates, the amp sounds fine. Punchy, loud, articulate, etc. just like one would expect. In those cases, the power supply has time to recover in between notes.

Maybe Eden could remarket the E/I core version and call it "the one with more tube rectifier style sag".

Cheers! Appreciate your work.
CBG