Eva, if you want to see some flyback insanity have a look at this lab power supply from HP:
http://cp.literature.agilent.com/litweb/pdf/5959-3344.pdf
They use a big asymmetric half bridge 1kW flyback
to get constant power over a large part of the output voltage range. Didn't Labgruppen use flyback at some insane power level too? Does anyone know what primary connection they used? It feels like the leakage spike will be a big problem if not using active clamping or an asymmetric half bridge connection.
But I guess you were thinking of the discontinous input current (which is discontinous even in continous mode) causing filtering problems at higher powers for a flyback PFC?
http://cp.literature.agilent.com/litweb/pdf/5959-3344.pdf
They use a big asymmetric half bridge 1kW flyback
to get constant power over a large part of the output voltage range. Didn't Labgruppen use flyback at some insane power level too? Does anyone know what primary connection they used? It feels like the leakage spike will be a big problem if not using active clamping or an asymmetric half bridge connection.But I guess you were thinking of the discontinous input current (which is discontinous even in continous mode) causing filtering problems at higher powers for a flyback PFC?
megajocke said:Didn't Labgruppen use flyback at some insane power level too? Does anyone know what primary connection they used? It feels like the leakage spike will be a big problem if not using active clamping or an asymmetric half bridge connection.
But I guess you were thinking of the discontinous input current (which is discontinous even in continous mode) causing filtering problems at higher powers for a flyback PFC?
You can make nice and simple active clamps with Power Integrations LNKxxx ICs. For example, you can take energy from a 580V "clamp" rail and return it to the 350V main rail. The "clamp" rail may be kept regulated exactly at 580V, just below IGBT/MOSFET breakdown. I'm using that approach in my high power PFC and high voltage class D circuits for ultimate efficiency.
Most of these ICs are intended for flyback operation, but there are some tricks to make them work in buck mode. Then no auxiliary transformer is required, only a few off-the-shelf axial inductors connected in series.
BTW: Yes, I was talking about high discontinous input currents in PFC requiring additional EMI filtering and causing unexpected transformer/inductor losses.
It's better as a liniear transformer. You are free to add as much as you like extra capacitors on the output.
The UcD amp has a very good PSRR, it has no sense to design a SMPS with a very low rimple voltage at rated output power.
It's more importend to have the voltage regulated, so the amp will not run on too high voltage when you have a high mains voltage.
Jan-Peter
The UcD amp has a very good PSRR, it has no sense to design a SMPS with a very low rimple voltage at rated output power.
It's more importend to have the voltage regulated, so the amp will not run on too high voltage when you have a high mains voltage.
Jan-Peter
Thanks for your answers Jan-Peter:
Good line/load regulation and low noise/ripple is excellent.
Is there a way to modify this design to remove the ripple??
Or can anyone recommend anything else?
Well, I wanted a very low noise and low ripple 0-~20kHz SMPS to be able to use small caps. I do not like big caps.Jan-Peter said:
I'm sure the SMPS+UcD will work well but even with high PSRR low PS noise/ripple is important...to me. I want to use a SMPS with analog amps.Jan-Peter said:
Yeah, good line and load regulation is good.Jan-Peter said:
Good line/load regulation and low noise/ripple is excellent.
Will line/load regulation and noise/ripple be included in the final datasheet?Jan-Peter said:
Is there a way to modify this design to remove the ripple??
Or can anyone recommend anything else?
It is not possible to remove the ripple except by using much larger caps on the output or a post-regulator. But the ripple is much more benign with less high audio freqency components than the sawtoothy-triangle wave from a standard line freqency power supply. Unless the amp has really really poor PSRR it shouldn't really be a problem.
Eva: That's quite smart, it should have very low component count and size! When misused as buck regulators these should be able to push quite a bit more power than what they are rated for in the standard discontinous flyback circuit...
Are you using this in the PFC with a magnetic snubber? I remember a circuit you posted once with a secondary winding on the snubber inductor and some diodes for clamping. What's the benefeit of doing it with a clamping rail? Leakage inductance in snubber inductor? Or are you sharing it between PFC and amplifier?
Eva: That's quite smart, it should have very low component count and size! When misused as buck regulators these should be able to push quite a bit more power than what they are rated for in the standard discontinous flyback circuit...
Are you using this in the PFC with a magnetic snubber? I remember a circuit you posted once with a secondary winding on the snubber inductor and some diodes for clamping. What's the benefeit of doing it with a clamping rail? Leakage inductance in snubber inductor? Or are you sharing it between PFC and amplifier?
domcheng said:
The reason is to have an regulated output voltage in dependable of mains voltage. A nice side effect is that we do not need a 115V/230V switch anymore, the SMPS will keep the voltage stable between 90V and 250VAC.
Actually the PFC function is used to regulate the output voltage.
Jan-Peter
Hi Jan-Peter,
I'm thinking of building a UcD-amp to replace my Arcam A90 (monitors: PMC-OB1). A few questions:
- when will smps for UcD400 and 700 be available, and what will be the price?
- Do i need softstart when using smps?
- Which UcD (180/400/700) has best audiophile quality for use in livingroom?
Thanks,
D
I'm thinking of building a UcD-amp to replace my Arcam A90 (monitors: PMC-OB1). A few questions:
- when will smps for UcD400 and 700 be available, and what will be the price?
- Do i need softstart when using smps?
- Which UcD (180/400/700) has best audiophile quality for use in livingroom?
Thanks,
D
Douwe01nl said:
Hi Douwe,
There are now prices yet avalaible for an SMPS UcD400 or even UcD700. we have just started with the design of the 360W SMPS, the >1kW is scheduld for later this year.
You will not need a soft start anymore.
Alle amps has more of less the same sonic signature, altough some customers claims the UcD700HG with HxR do sound very, very good.
Regards,
Jan-Peter
Re: how much more do we have to wait?
Hi Peter,
We have received our pre-production batch of 30 pcs and I can say that the SMPS is performing very well!
Nice stable, independable of the mains voltage you will have a constant output voltage. Very low EMI, it will meet the official standard with a nice margin.
We still have to do a couple of more test, but it seems that we can go in production in about 7-10 days!
So...if everything goes as planned we will have the 180W SMPS on stock in about 6 weeks.
Price will be € 125,00 ex. VAT and ex. works.
Many thanks to all our customers for their patience....
Next week I will post some pictures.
Regards,
Jan-Peter
peterhenk said:
Hi Peter,
We have received our pre-production batch of 30 pcs and I can say that the SMPS is performing very well!
Nice stable, independable of the mains voltage you will have a constant output voltage. Very low EMI, it will meet the official standard with a nice margin.
We still have to do a couple of more test, but it seems that we can go in production in about 7-10 days!
So...if everything goes as planned we will have the 180W SMPS on stock in about 6 weeks.
Price will be € 125,00 ex. VAT and ex. works.
Many thanks to all our customers for their patience....
Next week I will post some pictures.
Regards,
Jan-Peter
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.