Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Deciding flyback converter IC

Status
Not open for further replies.

yanyan999

Newbie level 3
Joined
Feb 6, 2018
Messages
3
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
42
Hello,

I'm trying to build an offline converter that has an input 150Vac-250Vac and outputs 24V 0.3A (about 7.2W). This converter will be used as an adapter to feed other digital and analog circuitry components on a PCB and is not used as a charger.

I have TI's ucc28911 (CV/CC controller) and Power Intergration's LNK626PG (CV controller) in mind.
Both are primary side regulator flyback ICs, are opto-less and can provide the output power required.

Both claim to be great solutions however TI comes with more protection such as input undervoltage, output overtalge and output overcurrent which are not present in LNK626PG. PI claims to have better EMI reduction based on E-Shield™ transformer construction techniques present in their PI Expert design software.

When I was trying PI Expert's suite and TI's Webench suite I found PI Expert's suite to more robust and detailed compared to TI's Webench and this a major reason that attracts me to PI's LNK626PG.

Has anyone tried any of the ICs mentioned? What about the design suites? Can someone fully rely on the outputs of the designer suites to create circuitry?
How should I make up my mind in deciding the IC?

Thank you so much
 

One way to evaluate would be total cost. Usually cost is a major deal breaker.

Having said that, when you are in exploratory mode, it is wise to make a list of your required performance parameters, and divide them in two, those must-have, and those nice-to-have.
Compare both. You may have to build prototypes and actually perform some lab work.
 

One way to evaluate would be total cost. Usually cost is a major deal breaker.

Having said that, when you are in exploratory mode, it is wise to make a list of your required performance parameters, and divide them in two, those must-have, and those nice-to-have.
Compare both. You may have to build prototypes and actually perform some lab work.

Thank you for your reply. From the cost perspective they both have the same price. Have you designed a flyback power supply before?
 

I have already done a project based on TI Webench with another IC (I don't remember now what device), which in the end, had to undergo some adjustments; I don't know if this was a bug and/or if it has already been fixed, but whenever possible give some margin of tolerance in the input specification of the project using this tool, higher than it really is.

Regarding your project, specifically speaking, in addition to the cost factor mentioned above, I would also add the vendor stock aspect. It is frustrating if have to redesign a footprint/PCB when we realize that a particular component (especially inductors/transformers) are not easily available to buy.
 
I have already done a project based on TI Webench with another IC (I don't remember now what device), which in the end, had to undergo some adjustments; I don't know if this was a bug and/or if it has already been fixed, but whenever possible give some margin of tolerance in the input specification of the project using this tool, higher than it really is.

Regarding your project, specifically speaking, in addition to the cost factor mentioned above, I would also add the vendor stock aspect. It is frustrating if have to redesign a footprint/PCB when we realize that a particular component (especially inductors/transformers) are not easily available to buy.

Thank you dear andre_teprom for you insight.

I have another set of questions:

How do you usually assess if you have done a good job in your design? I mean after receiving the PCB and placing the components and running your circuit, what do you look for to say your flyback is doing a good job?

How do you make sure that the transformer that you ordered for prototyping complies with the specs you sent to the company? Because I'm trying to deal with companies that are not very popular just random companies I find on aliexpress.

Based on your expertise, when do you think a Y-type capacitor is needed between the primary and secondary?

Thank you very much
 

what do you look for to say your flyback is doing a good job?

How do you make sure that the transformer that you ordered for prototyping complies with the specs you sent to the company?

In my personal opinion, transformers for fly-back, generally speaking do not do a good job, as they usually have a gap in the core, which increases the spread of EMI (unless if you are using a Pot core, which confines emissions a few more), that is something more difficult to measure, but in other hands you can reduce its weigh a few. In shorts, you are the one which have to rank the most important requirements, such as cost, efficiency, etc...
 
All transformers have primary to secondary stray capacitance. This will allow some higher frequency components to leak thru and create EMI issues.

The Y cap shunts those components back.

Whether you require it or not, the regulatory compliance testing will tell you so.
It is a wise idea to leave a placeholder on the PWB, to be able to fit a capacitor during debugging and testing.
 

From the cost perspective they both have the same price.

IC price is only one element of BOM cost, which is what
you (should) really care about.

An IC that requires more decoupling or high-end caps in
order to act right, or one that only approaches its advertised
performance with the most exotic magnetics, or needs a
6-layer PCB to get all the routes done, drives the BOM
cost out of proportion to its shelf price. You have alluded
to other concerns such as EMI which probably impact BOM
cost as well as maybe differentiating the two in dimensions
that aren't so immediately obvious.

If you live in (manufacture in) a locale with component
availability issues this may be even more of a problem, as
now it's not just about cost, but feasibility.

Might look and see if these two parts have vendor reference
designs and then drag a rake through the BOM and see how
the end solution stacks up.

Obtaining eval kits for both can let you see things that
Marketing doesn't want you to see.
 

Both of the IC's work OK if you follow the data sheets and build a transformer with reasonably low leakage inductance.

For really cheap, you can use a CMOS_555 to drive a low current 800V fet with simple opto coupled feedback...
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top