UCD180 questions

[Jaka Racman]

Hi Bruno,

4)PFC, if present, overshoots at around 17Hz (current feedforward never heard of, although many of these controller chips have dedicated input pins for this)

.

Well, PFC is designed for it. If you want to limit input current THD, then voltage loop must be slow or you are feeding second line harmonic into the input multiplier. Current feedforward is used only in B version of L4981 regulator of all the chips I am aware of. The same second you feedforward load current into input multiplier, THD will suffer also. And the main problem will be noncompliance with EN61000-3-3 (flicker) requirements. You might as well speed up the voltage loop and increase THD or use a rather large bank of capacitors. As far as i know, audio amplifiers are the last dinosaurs that are not subject to EMC harmonics and flicker standards. How do you do that? And why would you want to use PFC for power supply, when there is no requirement for that?

And BTW, output capacitor current control loop is used in SMPS design for some time now. ONsemi has 6 chips that use this technique, not counting older Cherry semiconductor ones.

Best regards,

Jaka Racman

 

[Bruno Putzeys]

Current feedforward is used only in B version of L4981 regulator of all the chips I am aware of.

Oops there you go... Extrapolation never works (it was this precise chip I'm using for my smps experiments so I supposed that current feedforward was commonplace)

The same second you feedforward load current into input multiplier, THD will suffer also. And the main problem will be noncompliance with EN61000-3-3 (flicker) requirements.

I don't exactly see why THD (actual harmonics) would increase, but surely the input waveform won't be a sinewave anymore. I was not aware of a thing like "flicker requirement".

As far as i know, audio amplifiers are the last dinosaurs that are not subject to EMC harmonics and flicker standards. How do you do that? And why would you want to use PFC for power supply, when there is no requirement for that?

IMHO audio amplifiers are exempt from these requirements precisely because it would force nearly the whole audio power amp industry out of business. Once a few consumer electronics companies have good PFC based power supplies you'll see it become a requirement rather quickly.

And BTW, output capacitor current control loop is used in SMPS design for some time now. ONsemi has 6 chips that use this technique, not counting older Cherry semiconductor ones.

As I said - everything is there for the right people to use, but when I look at the power supplies currently co-packaged with power amplifiers it's not happening (I haven't checked the supplies used by IcePower - presumably these are in a different league).

 

[Jaka Racman]

Hi Bruno,

don't exactly see why THD (actual harmonics) would increase, but surely the input waveform won't be a sinewave anymore.

Actually you are right. You have just found a way to push incredible amounts of non mains related harmonics into the mains network and not be punished for that. It is always good to have someone thinking out of the box. I actually wonder how much time will it take to the standardizing committees to fill this hole.

But the flicker problem still remains, and this will be a larger problem to fill if they impose this standard on audio industry like they did on mine (medical field). Because of that standard I am currently designing a flicker free power supply which will have more than 3.7kJ stored energy in capacitors additional to current feedforward. Some of our competitors are forced to specify dedicated power lines because of that. Audio industry must have a really good lobbyists.

Best regards,

Jaka Racman

 

[Ouroboros]

The 4981B simply has the ability to add a bit of spread-spectrum modulation to the oscillator to lower the conducted noise when measuring with an average responding EMC receiver.

I've designed in the 4981 as the PFC stage in a 250W PSU here at work. The chip works very well and nothing has blown up yet on the prototype! (The main isolating stage that follows the PFC is a two-transistor forward convertor producing 46V at 5A)

The site http://henry.fbe.fh-darmstadt.de/smps_e/#smps is extremely useful to help design and simulate various power conversion stages.

 

[Bruno Putzeys]

Actually you are right. You have just found a way to push incredible amounts of non mains related harmonics into the mains network and not be punished for that. It is always good to have someone thinking out of the box. I actually wonder how much time will it take to the standardizing committees to fill this hole.

Well it'll always remain something of a balancing act between energy storage and mains pollution. However the situation, I think using a pfc even with heavy current ff will still be much preferrable over a straight cap-loaded rectifier.

Also the non-harmonic components are not going to contribute to distribution losses (which is why we want good power factors) and the requirements don't need to be strict because the additional rubbish is asynchronous/noncorrelated so over a large number of stereo sets it'll average out (unless everyone in the country is playing the same radio station at full blast). Harmonics, being synchronous, do cumulate, so strict regulation is necessary there.

I'm not sure if audio has a such a good lobby. It's more likely we're quantité négligeable.

 

[Jan-Peter]

Guys,

Perhaps a good moment to start a new thread about how to design a good SMPS for an amplifier, analog or Class-D.

In this way we keep the thread about the UcD180 more clean to the subject.

Cheers,

Jan-Peter

www.hypex.nl

 

[Hans L]

Hi Xman737,

I just send you an email about details of ordering. Within a few weeks we will have a webpage ready were everybody can order the Class-D modules, together with cable assemblies and powersupplies.
...
Best regards,

Jan-Peter

www.hypex.nl

I did it... I read 18 pages of posts and feel like

Anyway, is the powersupply already available? I hope I'm not too impatient... I am however a complete twit when it comes to electronics and wouldn't want to buy a kit that isn't foolproof.

Does it need a soft-start circuit to be able to use it for tweeters in an active design? I noticed 'Pop-free stop mode' in the specs, is that enough? I saw someone stating it wasn't necessary in an active design, but another builder did have a softstart in his amp most likely designed for a passive speaker...

Thanks!
Hans.


P.S. I'm also looking forward to a paper with the answers to these questions by wytco0

Things that I think would be useful
Description of power supplies recommended, possibly some actual designs, eg dual mono, single; tranformer ratings etc
Description of signal connection, balanced vs. unbalanced.
Protection information, Switch on etc.
Cooling recommendations.
Opamp discussion
Info re cooling requirements, eg making it clear that although they run cool some heatsinking is needed.

 

[classd4sure]

Hi,

You only really require a soft start circuit with huge toroidal transformers as they induce a good surge of current when you first power them up, also large capacitor banks coupled to said transformer make it worse.

At what point does soft start become manditory? 300VA? 500VA?

I also would like to know if Hypex has plans for offering soft start circuits?

I've read on some sites who do sell them that they improve the sound by starting it without a punch of current and reduces "warm up" time by hours before it starts to sound good... true or false? ?

Thanks

 

[Bruno Putzeys]

You only really require a soft start circuit with huge toroidal transformers as they induce a good surge of current when you first power them up, also large capacitor banks coupled to said transformer make it worse.

At what point does soft start become manditory? 300VA? 500VA?

I also would like to know if Hypex has plans for offering soft start circuits?

I've read on some sites who do sell them that they improve the sound by starting it without a punch of current and reduces "warm up" time by hours before it starts to sound good... true or false? ?

Yup, soft start is a courtesy to your rectifier diodes and the circuit breakers in your house, but the amp doesn't need it.
If you read through the thread, you will find posts on power supply dimensioning, and how the amount of storage capacitance depends only on rated load impedance, not rated power. Once you got that, you'll see it is unnecessary to stack up such an amount of capacitance as to necessitate a softstarter.

IMHO soft start has no effect on warmup time. "Warmup" (as defined as the slow change in sound, usually for the better, during a certain time after powerup) is an electrochemical affair in the electrolytics. If you want to reduce warmup time, keep the power supply up all the time and turn the amp on/off by the enable pin (if at all you would bother turning it off).

 

[classd4sure]

Yeah I saw that as marketing hype but thought I'd ask

So if the reservoir capacitance isn't a problem, or, much of the problem, at what point does the transformer size necessitate soft start, given the standard 25 amp breaker?

From http://www.powertronix.com/html/body_in-rush.html
Transformer Rating
Recommended Protection

15 VA - 300 VA
None

300 VA - 1.0 kVA
Slow blow fuse

1.0 kVA - 2.0 kVA
Small external resistor in the primary circuit

2.0 kVA - 10.0 kVA
NTC thermistor or soft start circuitry

 

[Hans L]

Can you guarantee that my tweeter won't go plop! one last time while turning the amp off or on (what is more likely to cause damage: on or off)? What size is the thansformer that Hypex is going to offer with the UCD180?

Thanks,
Hans

 

[Jan-Peter]

Hans,

What you hear in your tweeter when you switch the amp on is a VERY small thick. Switching the amp off, doesn't create any noise........

When the customs has FINALLY gives our transformer free we will have a 160VA o-shape toroidel transformer.

Cheers,

Jan-Peter

www.hypex.nl

 

[peranders]

At what point does soft start become manditory? 300VA? 500VA?

I also would like to know if Hypex has plans for offering soft start circuits?

I say the limit is around 300 VA if you have 6 A fuses in the wall. 600 VA and 6 A fast did blow my fuses one time of ten. With a slow fuse it's OK with 6 A but the mains switch takes abuse. I'd say there is no problems with the rectifiers though.

Lars Clausen, Rod Elliott and I have ready solutions for soft start. Rod's design uses tranformer but both my and Lars' solutions are "hot", floating at mains potential.

 

[pa]

Batteries and charger(s)

Hi,
what do you think about using batteries (with a regulated charger wich could work in Switch mode) as capacitors this could be coupled with good capacitors for quick peaks ?

 

[wytco0]

I have a one of Rods soft start boards on its way to me from my favorite place, Australia.

As a not very technical beginner the transformer solution looked safest to me. I may not need it for this project (not sure what transformers I will use) but I thought the board was worth having anyway.

Rod's stuff seems well documented, l and I know he responds to emails quickly.

 

[matjans]

psu probs

hi all,

i have a small problem with my psu ( i guess). When i got my ucd modules i quickly threw a psu together. It consisted of an ilp ~500va 4x15v toroid (of which i wired 2x2 secondaries in series to get me 30vac secondaries, or ~42vdc), 1x jamicon 4700uF 50V and 1x philips 1000uF 50v caps per rail. Worked perfectly, no noise, no rumble, good sound. The toroid i borrowed from a friend of mine.

After receiving some components from RS and a deep dive in my parts-box i built a new psu: amplimo dual 33v secondaries, 4x low impedance panasonic FC 2200uF 50V, 1x47uF black gate std, 1x4,7uF 50V black gate N (not necessary, i know but i had them laying around and i figured it wouldn't do any harm) per rail.

Problem: noise and hum. High frequency noise in my tweeter and a (i guess, no scope on hand) 50hz rumble in my woofers. SOunds better than the previous psu though. (slightly more transparent and even more bass control fyi)

Does anyone know or can anyone guess what's going on here ?

 

[Bruno Putzeys]

Re: psu probs

Problem: noise and hum. High frequency noise in my tweeter and a (i guess, no scope on hand) 50hz rumble in my woofers. SOunds better than the previous psu though. (slightly more transparent and even more bass control fyi)

It appears that you changed other connections too, not just the power supplies. See if the wiring of the two versions are identical.

 

[matjans]

nah, prolem wasn't the psu, it was my preamp.

 

[nickywicky]

ok to use 3 Bridges?

I'm planning on getting 3 UCD400's when they become available and I'm sorting out the PSU in the meantime.

Found a really good deal on a 600VA 40-0-40 toroidal.
( 1/2 the price of a 500 va 35-0-35 if they even had
35vac ones! )

Problem is the no load voltage will be 61.5vdc. ( say 6% regulation of t/former at no load , voltage on New Zealand is always 4% higher than the rated 230 * 1.4 and -0.7 v for
the bridge), and the UCD400 is 60V max.

My question is, is it a reasonable to use a further 2 bridges to subtract 1.4Vdc to bring me pretty close to 60Vdc ?

 

[Bruno Putzeys]

So you're basically asking if it's OK to place additional diodes in series with the rectifier in order to drop n*0.6V.
If that's what it takes to save you the cost of having to buy a new transformer it is OK of course. I don't feel too comfy about running the amp at +/-60V (idle) though, because you can't be too sure of the mains voltage. It's best to have a safety margin such that only at 10% overvoltage on the mains the idle voltage actually hits +/-60

So if you have 4% overvoltage, try to go for 56V (idle).

That you have 4% overvoltage is not an accident btw. This is a leftover from times when the nominal mains voltage was spec'ed at 240V. They now specify 230V but with different tolerances such as to cover the same voltage range as before