I want to convert ~5.4Vdc to ~50Vdc @ 200mA.
I've chosen to use a forward converter topology, but there are soooo many controllers available.
The output needs to be isolated, so the controller must be of the push pull, dual transistor kind.
I would like to controller to be as simple, small, reasonably efficient and not too expensive; but since they only cost a few dollars each, cost isn't such a major problem.
I wanted to ask you guys who've made PWM supplies before about this as this is the first I'll have made and I'm not familiar with which manufacturer's offer such options.
A lot of the newer controllers I'm seeing are designed for single transistor clamp operation that won't be acceptable in this case.
Syncronised switching on the output would help with efficiency, but again, I can't really allow this converter's complexity to get out of hand.
Please feel free to suggest any particular method you think might be usable!
I posted something similar to this recently, then I discovered that forward converters are just yet another thousand pages of reading on standard boost converters!
I've chosen to use a forward converter topology, but there are soooo many controllers available.
The output needs to be isolated, so the controller must be of the push pull, dual transistor kind.
I would like to controller to be as simple, small, reasonably efficient and not too expensive; but since they only cost a few dollars each, cost isn't such a major problem.
I wanted to ask you guys who've made PWM supplies before about this as this is the first I'll have made and I'm not familiar with which manufacturer's offer such options.
A lot of the newer controllers I'm seeing are designed for single transistor clamp operation that won't be acceptable in this case.
Syncronised switching on the output would help with efficiency, but again, I can't really allow this converter's complexity to get out of hand.
Please feel free to suggest any particular method you think might be usable!
I posted something similar to this recently, then I discovered that forward converters are just yet another thousand pages of reading on standard boost converters!
If you want to "up-convert" and having isolated supply then I suggest you to look at the Flyback topology also.
Then you need only one transistor and no clamping issues and no output flyback inductor, an experienced designer can also with this topology sense the secondary voltage without having any electronics connected to the secondary but i don't suggest you go that far if your experiences with SMPS is very limited as you mentioned.
Then you need only one transistor and no clamping issues and no output flyback inductor, an experienced designer can also with this topology sense the secondary voltage without having any electronics connected to the secondary but i don't suggest you go that far if your experiences with SMPS is very limited as you mentioned.
Hi Jaka,
Thanks for the lead. I checked the 845 out. It looks really good and it's low output voltage wouldn't be a problem with a step up transformer, but it's output is a bit on the low side.
There is a similar version made by Maxim that outputs 1W, as opposed to the 785mW of the 845. But I could reall use something closer to 5 - 10W; with 10 being perfect.
Do you know of any similar controllers with higher power outputs?
Thanks for the help!
I had initially avoided the flyback method when the first few things I read about SMPS mentioned that the flyback design lead to much lower output supplies.
It suggested that because a flyback design only uses on polarity of magnetic flux, that this would cause the transformer to saturate at lower power ratinings, meaning an increased physical volume of the supply.
Obviously, you know a lot more about this than me, so I'd really appreciate any futher help with this.
I've spent many hours looking through PDF's already, and my mind is approaching the melt-down stage. Looking at the PDF's though, the newer single transistor controllers seem to look a bit like flyback designs, but they have complex looking magnetics, involving clamping coils and flyback windings on the outputs.
If I could find a higher rated version of the 845, that would be excellent!
Again, thanks for your help everyone!
A possible match?
The 1425 also may be a possibility, although I think that it's output only reaches 6W.
Is this the smart sensing topology you mentioned Ultima? It mentions sensing transconductance?
The 1425 also may be a possibility, although I think that it's output only reaches 6W.
Is this the smart sensing topology you mentioned Ultima? It mentions sensing transconductance?
I built something similar (5V -> 48V @ 100mA) a long time ago.
I used a MAX1771 directly driving a IRLZ44N MOSFET, and used a flyback transformer wound on a gapped ETD29 core. I wasn't using isolation, but you could probably do something creative with an optoisolator to accomplish this. Also, the MOSFET wasn't suitable at all for the application, really... you could probably find a much better one with a lower Rds(on) and a lower Vds(max).
I used a MAX1771 directly driving a IRLZ44N MOSFET, and used a flyback transformer wound on a gapped ETD29 core. I wasn't using isolation, but you could probably do something creative with an optoisolator to accomplish this. Also, the MOSFET wasn't suitable at all for the application, really... you could probably find a much better one with a lower Rds(on) and a lower Vds(max).
eeka chu said:
Hi eeka,
you seem to have found a nice regulator that could be used.
I also notice that they have incorporated a smart sensing of the flyback voltage seen on the secondary, it's in the first picture and it's around Q1 and R2.
Don't worry about those issues you mentioned, flybacks are used in power supplys up to even 500 watts and can still be small.
The thing with flybacks is that you need a core with an airgap on the center legg, that ensures the core not to saturate, or let us say it moves up the limit.
Now you need to find a suitable core and start to calculate, you say 50V @ 200 mA, that's not very much.
qmarsh mentioned that he used an ETD 29, I'm sure you will be fine with much smaller core for your max 10-12 watts.
You say that you have a 5,4 V source, do you know the lowest voltage seen from that source if you load it with 12 watts??
Just in case it shouldn't go below 4,5 V since that was the lowest supply voltage for that LT IC you have found.
Good luck!
Thanks Ultima,
I've been reading through the datasheet for the controller, but a couple of things are still a bit unclear.
When I looked at inductor based converters, the inductor had to be quite specifically chosen to get the best from the converter - to balance it's inductance with it's peak current rating and it's DC resistance.
I read through the transformer section of the datsheet, but it doesn't seem to mention optimising the transformer's inductance, or anything in particular about peak current ratings.
Before I decided to go with the forward / flyback topology, I had been working through an inductor based supply. And so, I now have five sample MOSFET's from Fairchild. These are their from their Powertrench line, and so they're specifically suitable for DC converters. They have a very low gate charge, are rated for 10 amps and brake down at 60Vdc.
I thought it would be good to put these to use in the flyback supply, but I'm not so sure about the flyback voltage.
I've been reading through the datasheet for the controller, but a couple of things are still a bit unclear.
When I looked at inductor based converters, the inductor had to be quite specifically chosen to get the best from the converter - to balance it's inductance with it's peak current rating and it's DC resistance.
I read through the transformer section of the datsheet, but it doesn't seem to mention optimising the transformer's inductance, or anything in particular about peak current ratings.
Before I decided to go with the forward / flyback topology, I had been working through an inductor based supply. And so, I now have five sample MOSFET's from Fairchild. These are their from their Powertrench line, and so they're specifically suitable for DC converters. They have a very low gate charge, are rated for 10 amps and brake down at 60Vdc.
I thought it would be good to put these to use in the flyback supply, but I'm not so sure about the flyback voltage.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.