Emotiva on transformers vs SMPS

[MZKM]

Nice to see they have an Audio Precision; now if only they followed Schiit & Topping in releasing the AP reports.

 

[gfinlays]

Hehe, properly debunks the audiophile myth that expensive LPS are better (lower noise) than SMPS.

 

[muslhead]

I had to return my AVR last week to Emotiva. Turned out the problem was faulty caps on the PS board. Measurements are one thing, reliability is another.

 

[gene_stl]

Linear power supplies can be made to be quiet and they are repairable and protect the gear from line disturbances more than most SMPS.
SMPS often have custom chips and can't be field repaired. No thank you. They also may have less power line isolation which makes the green ground connection more important.

 

[gfinlays]

I had to return my AVR last week to Emotiva. Turned out the problem was faulty caps on the PS board. Measurements are one thing, reliability is another.

To be fair, that's really the cap manufacturer's issue. I built a pair of power amps a good few years back - a week after completing them, one of the Black Gate input caps went bad. The amp would distort really badly for a few seconds after power up and then settle down OK. After replacing the cap all was well again.

 

[muslhead]

To be fair, that's really the cap manufacturer's issue. I built a pair of power amps a good few years back - a week after completing them, one of the Black Gate input caps went bad. The amp would distort really badly for a few seconds after power up and then settle down OK. After replacing the cap all was well again.

Maybe, maybe not. Could be Emotiva for cheapening out on the cap quality to keep costs down too. Either way i wasnt blaming Emotiva and wouldnt until I knew for sure what quality caps they used.

 

[MakeMineVinyl]

Just an obversion - you will note that the 2500 VA transformer actually can take up equal or less space in the chassis (depending on orientation) than the SMPS of equivalent power. So at least as far as space saving, they are essentially a wash. I'm assuming the filter capacitors are separate on each channel module as is widely done. Obviously the transformer is far heavier, so I'm not arguing that point.

Both linear and SMPS supplies have their strengths and weaknesses - an SMPS does not make a linear supply obsolete.

 

[chris719]

Linear power supplies can be made to be quiet and they are repairable and protect the gear from line disturbances more than most SMPS.
SMPS often have custom chips and can't be field repaired. No thank you. They also may have less power line isolation which makes the green ground connection more important.

High isolation is a feature of many SMPS...

 

[Deleted member 21219]

Measurements are one thing, reliability is another.

Hear, hear! You can't listen to a system that doesn't work!
Jim

 

[wwenze]

Just an obversion - you will note that the 2500 VA transformer actually can take up equal or less space in the chassis (depending on orientation) than the SMPS of equivalent power. So at least as far as space saving, they are essentially a wash. I'm assuming the filter capacitors are separate on each channel module as is widely done. Obviously the transformer is far heavier, so I'm not arguing that point.

Both linear and SMPS supplies have their strengths and weaknesses - an SMPS does not make a linear supply obsolete.

Hmm....

https://sg.rs-online.com/web/p/toroidal-transformers/1176084/?cm_mmc=SG-PLA-DS3A-_-google-_-PLA_SG_EN_Power_Supplies_&_Transformers_Whoop-_-(SG:Whoop!)+Toroidal+Transformers-_-1176084&matchtype=&pla-332939895043&gclid=Cj0KCQjwo-aCBhC-ARIsAAkNQisoOBOai8hZuBBC3tHVQ2iHwIl_IYdyP4GIafwNVqv08fRPA4HOAxUaAq5NEALw_wcB&gclsrc=aw.ds
2000VA, 238mm diameter x 75mm height


https://www.fsplifestyle.com/PROP182003192/
2000 watts, 200 X 150 X 86mm including capacitors and rectifiers (or secondary side switching elements), and regulated


The SMPS module in the video has a bigger area in two dimensions yes... but total volume, not so.

 

[restorer-john]

Hehe, properly debunks the audiophile myth that expensive LPS are better (lower noise) than SMPS.

Doesn't "debunk" a thing actually. He shows nothing useful in that self serving video. An idle noise plot with no conditions stated or shown? No signal, no performance comparisons etc. Yawn.

The "bottom line" main reason is of course cost savings- something he conveniently forgot to mention. Cost in shipping and cost in txformers and large capacitors.

Reliability goes out the window too.

 

[MakeMineVinyl]

The SMPS module in the video has a bigger area in two dimensions yes... but total volume, not so.

As a manufacturer, we ran into this tradeoff. We tried to incorporate a SMPS, but ultimately we went with transformers because they took up less space in an already crowded chassis. The form factor if SMPS is kind of unwieldy in some circumstances, especially when heat sinking requirements are taken into account.

 

[MakeMineVinyl]

Reliability goes out the window too.

In my experience, a transformer will last essentially forever - a SMPS, not so much. More to the point, I have vintage gear which is as much as 60 yeans old, and have never had a transformer issue. As a matter of fact, I don't think I've ever had to replace a transformer in any piece of equipment. Over the decades, I've replaced many, many SMPSs though; some have 'let go' quite spectacularly.

 

[Mike-48]

It's bizarre to compare a transformer to an SMPS. The transformer needs rectifiers and large capacitors to make a power supply. The SMPS is complete. Its capacitors can be smaller because the filtering frequency is less. And while transformers never fail, electrolytic capacitors fail regularly.

 

[wwenze]

And while transformers never fail

Does this count lol

 

[gene_stl]

It is true that there are medical grade SMPS that have low leakage currents.
I have also replaced many sick power transformers in linear supplies.
But by comparison I have had to deal with a higher rate of SMPS failures since they were introduced.
restorer-john's observation about cost and schipping weight is the same as my observation. It's all about saving money and maybe efficiency ratings when new for the government certifications. If it's up to me I prefer linear. I realize that like LED lighting SMPS are an idea whose time has come.

 

[MakeMineVinyl]

, electrolytic capacitors fail regularly.

I guess the filter capacitors in my decades old power amps never got that memo.

 

[chris719]

Reliability goes out the window too.

What components fail in SMPS? The same ones that fail in linear power supplies. Maybe you get the odd controller failure, but it’s not as common. I have seen 30 year old medical devices in the field with SMPS, so let’s not get too crazy. I have a 6 GHz Tek spectrum analyzer that runs Windows coming up on 25 years old that has the original SMPS too.

If you design with cheap caps or don’t pay attention to thermals and ripple current they will die, but the phenomenon is not entirely unique to SMPS. If linear PSUs were still attractive for commercial gear there would be tons of cheap crappy linear supplies out there with bad caps as well.

 

[chris719]

I guess the filter capacitors in my decades old power amps never got that memo.

Five cent knockoff caps didn’t exist when your power amp was made. I don’t see too many dead Panasonic, Nichicon, Fujitsu, etc. caps in any SMPS.

 

[chris719]

As a manufacturer, we ran into this tradeoff. We tried to incorporate a SMPS, but ultimately we went with transformers because they took up less space in an already crowded chassis. The form factor if SMPS is kind of unwieldy in some circumstances, especially when heat sinking requirements are taken into account.

A SMPS can be made smaller for the equivalent power. It’s not even a question really. Look at the size of some USB-C chargers. The efficiency is super high on state of the art supplies. It’s not really fair to compare regulated to unregulated anyway.


https://www.ti.com/tool/PMP20637