Stereophile has started calculating EPDR.

[amirm]

Well said. No way you want to compromise in performance of speakers for EPDR.

 

[Sokel]

I'm going to add this comment I've just put in another thread where EPDR was discussed.

Further - EPDR is becoming increasingly irrelevant - as more and more speakers are driven by Class D amplifiers. In fact if you are worried about it - the best solution (IMO) is not to compromise your speaker selection to exclude those with non-benign phase shifts - but to use a Class D amp.

*******
I also particularly dislike EPDR as a concept. The way it is framed as a lower impedance is particularly misleading IMO. It is not. EPDR lower than the actual impedance doesn't require higher current from an amp.

And it is almost useless in understanding how any particular amp will respond to any given speaker load. In fact if you read the AES paper - it is more framed as a guide for designers to better understand how to dimension the output stage power loss requirements for Class B or Class AB designs)

IMO it doesn't help consumers at all, at least as far as amp selection is concerned for any given speaker. It might help in selecting speakers that are easy to drive. But then only if you are planning to drive them with a Class AB amp.

Judging by the fact that half of the top THD+N champions at the ASR list are AB class amps, EPDR is not irrelevant.

People do consider the list as buying guide and combined with the fact that they are small amps in general, one has to take advise on them.

 

[amirm]

Judging by the fact that half of the top THD+N champions at the ASR list are AB class amps, EPDR is not irrelevant.

Why? You have seen them get damaged? Or go into protection?

 

[Sokel]

Why? You have seen them get damaged? Or go into protection?

I never used one of them, so I wouldn't know.
Some show their limits at 2 Ohm or so.

Been on the (very) safe side of things as usual though, I (personally) wouldn't use them to drive arc-like loads.

 

[amirm]

Power is cheap these days. Get an amp with more wattage and there wouldn't be any concern. And as it happens, most of those that are rated highly by me are class D.

 

[antcollinet]

Judging by the fact that half of the top THD+N champions at the ASR list are AB class amps, EPDR is not irrelevant.

I said "increasingly irrelevant"

As is the relative performance of the best Class AB amps vs the best class D amps - what is relevant is the proportion of amps being bought that are now class D rather than class AB - and how that proportion is changing over time.

 

[valerianf]

Car audio speakers are going lower that 2 Ohms and need specifically designed amps to drive them.
Apart some very fancy high end speakers, I do not see the need to design a speaker with a very low EPDR.
It will require a new amp box capable of driving it, with no gain in the sound quality.
I remember an article about Ken Ishiwata saying that when working at Marantz they had to modify the amplifiers designs to make them compatible with a new crossover design from B&W.
I could add that it was obviously an under engineered design from B&W not taking care of the amplifier requirement.
EPDR curves are very useful to detect this awful behavior from speaker manufacturers: why buying a speaker that may damage a class AB amp?

 

[amirm]

EPDR curves are very useful to detect this awful behavior from speaker manufacturers: why buying a speaker that may damage a class AB amp?

Because they don't. Amps already have protection circuits for the SOA (Safe Operating Area). I have yet to hear someone say, regardless of what speaker they use, that cranking them up caused the amp to go into protection. Normally, the amp simply runs out of power.

SOA is also vastly improved in recent years so whatever the model uses, likely is way too conservative.

Emphasis on EPDR will force design changes for the worse. No reason to go there.

 

[valerianf]

Ken Ishiwata citation about B&W: "Unfortunately some of their speakers are very difficult to drive so when they introduced the original 801 and 800 we, Marantz Japan (a distributor of B&W) we had a lot of problems driving those speakers. Krell, Mark Levinson didn’t do the job. We studied the speakers and so we came up with the proposal of a different way of designing the power amplifier to drive and we made those monoblocks at that time."
When a speaker design makes obvious difficulties for renowned amplifiers brands, it is that there is an issue with the speaker design itself!
Let us not hide the real responsibility: no need of a class D (not EPDR limited) amp 2 Ohms capable to drive a well designed speaker in an usual room with a reasonable sound level.
In the past, many generations of audiophiles have made the demonstration.

 

[antcollinet]

no need of a class D (not EPDR limited) amp 2 Ohms capable to drive a well designed speaker

That could be turned around.


No need to compromise speaker design for EPDR concerns just so you can continue to use old tech class AB without thinking, when there is Class D available as a more modern and higher power for the money alternative.

 

[jackocleebrown]

Hi, just adding my 2 cents to this. I think that EPDR is useful for us as a loudspeaker manufacturer for a couple of reasons. We do get service cases from customers who complain that their amps go into protection when played moderately loud. I've also personally experienced a couple of such cases too at shows or during demos. In all cases this was with class-AB amplifiers, mostly with small form-factor amps (think all-in-one amp+streamers). In all cases the amp manufacturer blamed the loudspeakers for being "a difficult load", whereas I would put the issues down to either cost saving in the amplifier output stage, or poor SOA protection in the amplifier. EPDR was very useful in these cases to demonstrate that our speakers were not unusually difficult to drive, nor very different from others on the market. EPDR is also a very useful sense check that our crossover/speaker designs are not silly and likely to cause problems for some customers. Typically we try and keep EPDR > 2 Ohms. For high-end speakers it's not particularly problematic to go lower than this because you can almost guarantee that the customer will have plenty of power (although there are some tube amp fans out there).

 

[peng]

Hi, just adding my 2 cents to this. I think that EPDR is useful for us as a loudspeaker manufacturer for a couple of reasons. We do get service cases from customers who complain that their amps go into protection when played moderately loud. I've also personally experienced a couple of such cases too at shows or during demos. In all cases this was with class-AB amplifiers, mostly with small form-factor amps (think all-in-one amp+streamers). In all cases the amp manufacturer blamed the loudspeakers for being "a difficult load", whereas I would put the issues down to either cost saving in the amplifier output stage, or poor SOA protection in the amplifier. EPDR was very useful in these cases to demonstrate that our speakers were not unusually difficult to drive, nor very different from others on the market. EPDR is also a very useful sense check that our crossover/speaker designs are not silly and likely to cause problems for some customers. Typically we try and keep EPDR > 2 Ohms. For high-end speakers it's not particularly problematic to go lower than this because you can almost guarantee that the customer will have plenty of power (although there are some tube amp fans out there).

IMO, it is useful, but unfortunately misleading by calling it because it seems that very few readers understand the EPD part, ended up focusing more on the R part and often mistaken it as an amplifier current capability issue vs the current demand by the speaker load. It would have been better if the originator had focused more on the heat, SOA related to the amp's output stage, rather than the R, ie "Resistance" that would inflate the calculated current that people would think in terms of Ohm's law, I see that happened even in the Benchmark article on their website, and in that case obviously J.S (an EE if I understood right) did so unintentionally.