toroid or normal E shaped hf transformers with 2 isolated bobbins are very safe for the smps's. Especially for the small switched mode power supplies you use for phones etc. They hardly produce any EMC.
For big smps's its much harder to meet the EMC specs.
For big smps's its much harder to meet the EMC specs.
Jan-Peter said:
Bosh bosh then.
If you want to plod in, as a sales bloke, and say your supply can be operated with or without and then have a blather about other stuff then 'all not well and good'.
If you want to back off and leave people wait for the proper answer then....
Bonus
DNA
The question of "when is it coming" has been loosely addressed so please refrain from off-topic arguing/speculation over the safety/performance of a product that hasn't even been produced yet.
Blarp!!!!
Are you after some ******** of the week award?
The basic rules, sometime in the past, meant people were not allowed to discuss 'dangerous' off-line stuff....
And now I'm having a rag at Jan for not demonstrating basic knowledge about such things.
Good job we are having an 'off topic' argument/speculation about the safety of his product before he has produced it.
He might produce something safe as a result.
Numpty.
DNA
Are you after some ******** of the week award?
The basic rules, sometime in the past, meant people were not allowed to discuss 'dangerous' off-line stuff....
And now I'm having a rag at Jan for not demonstrating basic knowledge about such things.
Good job we are having an 'off topic' argument/speculation about the safety of his product before he has produced it.
He might produce something safe as a result.
Numpty.
DNA
DNA, of course the "basic rules" still apply but this thread is about a commercial product not a DIY designed and built project. There's hardly the same risk factor involved.
Further, Jan-Peter posted to answer the questions of others, not to sell his products. By all means discuss the importance of double insulation, EMC requirements, creepage, etc. but refrain from personal attacks.
Further, Jan-Peter posted to answer the questions of others, not to sell his products. By all means discuss the importance of double insulation, EMC requirements, creepage, etc. but refrain from personal attacks.
Bgt said:
in that case I am pretty sure the RF radition exceeds the CE limits
that cap is placed for a reason, don't remove it
best
=
Guido
BWRX said:
I don't know what mushrooms you are on but Jan was/is responding to a query about the availability of an off-line Hypex SMPS for use by DIY people.
Saying There's hardly the same risk factor involved. does not cover my sausage with mustard.
Saying but refrain from personal attacks is like saying I was having a go without reason.
Hows about you get lost and get your head straight?
In the mean time Jan might have got lost and will be sorting out the stuff in his SMPS that means DIY people can slot one in without killing themselves if they RTM, Spare F.
DNA
DNA, I prefer beer but if I were on mushrooms I would have gotten them from you because you are on some of the best...
There is no argument that Jan was replying to a query about the availability of a Hypex SMPS. You were questioning the CONSTRUCTION of said SMPS which hasn't even been made available to the public. Where is the confusion?
You were having a go without reason. All Jan-Peter has presented so far is more or less marketing speak because they do not have a product that is ready to be released. There have been NO FORMAL SPECS published for said SMPS so there is no reason for you to attack him. Do you understand that?
Jan is not the one designing the SMPS, which was made clear from one of his previous posts.
There is no argument that Jan was replying to a query about the availability of a Hypex SMPS. You were questioning the CONSTRUCTION of said SMPS which hasn't even been made available to the public. Where is the confusion?
You were having a go without reason. All Jan-Peter has presented so far is more or less marketing speak because they do not have a product that is ready to be released. There have been NO FORMAL SPECS published for said SMPS so there is no reason for you to attack him. Do you understand that?
Jan is not the one designing the SMPS, which was made clear from one of his previous posts.
Gentlemen, sorry for interrupting the speculations but if Jan-Peter is still
awake, I would like to know how much gain in efficiency one can get by
using sync-rectification at such high output voltages.
In my world synchronized rectification starts paying off at about 15V and below for "normal" currents. Things may have moved forward though.
"btw the company I work at, stoole a power guy from a company Hypex work with"
/ BR / Mattias
awake, I would like to know how much gain in efficiency one can get by
using sync-rectification at such high output voltages.
In my world synchronized rectification starts paying off at about 15V and below for "normal" currents. Things may have moved forward though.
"btw the company I work at, stoole a power guy from a company Hypex work with"
/ BR / Mattias
Jan Peter,
One thing that you could reveal about your supply is perhaps some physical dimensions.
Like maximum building height including clearance from bottom if it comes as a bare board.
Or will you enclose it just leaving the output terminals for access with an IEC inlet.
I'm dreaming of something I can fit into a 1U rack.
If you can say something about the dimensions, then people can start
plan their housings.
Regards / Mattias
One thing that you could reveal about your supply is perhaps some physical dimensions.
Like maximum building height including clearance from bottom if it comes as a bare board.
Or will you enclose it just leaving the output terminals for access with an IEC inlet.
I'm dreaming of something I can fit into a 1U rack.
If you can say something about the dimensions, then people can start
plan their housings.
Regards / Mattias
Member
Joined 2003
As Jan-Peter revealed in this thread, the 180W supply will be first, and will fit a 1U rack enclosure.
See preliminary datasheet here.
See preliminary datasheet here.
ekaerin said:
Hi,
We will receive next week our pre-production batch....finally.... 😉
The height will be so low it will defiantly fit in 1HE, that's our goal!
The clearance/creepage between primary and secondairy will be 6mm, you will not need an earth conenction. A two pins inlet without earth will be sufficient.
We expect to have the production quantities in stock in about....6-10 weeks.
The eff. will be pretty good, the PFC will also have an active clamp.
Regards,
Jan-Peter
Cant wait.....
Dcibel, thanks for preliminary data sheet.
It's pretty good to reach some 90% eff. with a single stage converter having wide input.
No choke on the output.......
I'm not familiar with the topology but I think the sync rectifiers may pay
off due to the high reverse blocking needed ???
Looking fwd to see all skeptical faces until my next amp starts deliver.
/
😉 / Mattias
Dcibel, thanks for preliminary data sheet.
It's pretty good to reach some 90% eff. with a single stage converter having wide input.
No choke on the output.......
I'm not familiar with the topology but I think the sync rectifiers may pay
off due to the high reverse blocking needed ???
Looking fwd to see all skeptical faces until my next amp starts deliver.
/
😉 / Mattias
Synchronous rectification increases efficiency, particularly at low currents and even for +/-50V output. The usual challenge is gate drive and the poor man's approach is to use a couple of auxiliary transformer windings, but this is usually at the expense of current peaks and increased EMI when primary switches turn on and rectifiers turn-off. True synchronous rectification requires active control and some logic that is seldom used due to cost constraints.
The single stage SMPS with PFC is actually a flyback with discontinuous input current (thus not easily scalable to high powers) and produces substantial 100Hz output ripple by design.
The single stage SMPS with PFC is actually a flyback with discontinuous input current (thus not easily scalable to high powers) and produces substantial 100Hz output ripple by design.
Hi Eva,
I can't really follow your reasoning regarding the efficiency gain at low
currents, but never mind.
Yes, the 100Hz ripple must be taken care of but essentially the primary
bulk has moved to the secondary. Although lower voltage / energy you only need the lytics on the secondary.
And just because the topology is boost derived I believe you could use
anything from flyback to some type of fullbridge.
The active clamp Jan Peter mentioned will give high currents circulating
and with a little luck you could use these to reach ZVS.
But as I said I'm not at all familiar with this type converter, just gut feelings.
/ Mattias
I can't really follow your reasoning regarding the efficiency gain at low
currents, but never mind.
Yes, the 100Hz ripple must be taken care of but essentially the primary
bulk has moved to the secondary. Although lower voltage / energy you only need the lytics on the secondary.
And just because the topology is boost derived I believe you could use
anything from flyback to some type of fullbridge.
The active clamp Jan Peter mentioned will give high currents circulating
and with a little luck you could use these to reach ZVS.
But as I said I'm not at all familiar with this type converter, just gut feelings.
/ Mattias
Hi,
-we do not use synchronize regulation.
- we do use an active clamp circuit, to reduce the primary losses and to keep the SMPS in continues mode.
- it's no problem that there is still a ripple voltage, anyhow it's half of a normal transformer with the same capacitors. And the UcD has a good PSRR, sofar the SMPS and UcD works very good together.
- to further reduce the ripple voltage everybody is free to add more capacitors external.
In this paper we have explained more about SMPS and Class-D amps;
http://www.hypex.nl/docs/allamps hypex layout.pdf
Regards,
Jan-Peter
-we do not use synchronize regulation.
- we do use an active clamp circuit, to reduce the primary losses and to keep the SMPS in continues mode.
- it's no problem that there is still a ripple voltage, anyhow it's half of a normal transformer with the same capacitors. And the UcD has a good PSRR, sofar the SMPS and UcD works very good together.
- to further reduce the ripple voltage everybody is free to add more capacitors external.
In this paper we have explained more about SMPS and Class-D amps;
http://www.hypex.nl/docs/allamps hypex layout.pdf
Regards,
Jan-Peter
Take a 100V MOSFET with .020 ohm Rds-on and compare the voltage drop across the body diode for 2A with the voltage drop across Rds-on for the same current. Now compare the same figures at 20A.
A PFC requires a bulk inductor and in the PSU design that I saw the only big inductor is the own (gapped) transformer. That's why it has to be a flyback.
There are other methods to get "automatic" PFC from a flyback without 100Hz ripple on the output. See figure 15 on TDA16846-2 datasheet:
http://www.infineon.com/cms/en/serv...f24ec_fileId_db3a304412b407950112b41823ce24ed
A PFC requires a bulk inductor and in the PSU design that I saw the only big inductor is the own (gapped) transformer. That's why it has to be a flyback.
There are other methods to get "automatic" PFC from a flyback without 100Hz ripple on the output. See figure 15 on TDA16846-2 datasheet:
http://www.infineon.com/cms/en/serv...f24ec_fileId_db3a304412b407950112b41823ce24ed
Eva, I get it.
Unfortunately I have never used synchronized rectification although I have
considered it a few times.
The projects I usually work with don't have margins for "inventions".
That is, except for some bucks with Texas TPS40xxx controllers.
However nowadays there are ICs that may make it a bit easier to implement.
Regards / Mattias
Unfortunately I have never used synchronized rectification although I have
considered it a few times.
The projects I usually work with don't have margins for "inventions".
That is, except for some bucks with Texas TPS40xxx controllers.
However nowadays there are ICs that may make it a bit easier to implement.
Regards / Mattias
- Status
- Not open for further replies.