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.

Is 7 Mhz switching frequency too high for a boost? - Please Please Help.

Status
Not open for further replies.

drewcrew

Newbie level 6
Joined
Dec 12, 2011
Messages
14
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,397
Hello,
I'm trying to make a boost converter that will have a digital pwm control feedback that will be implemented with microcontroller. However, before I get started, I'm having a hard time figuring out the switching frequency.

My main requirements are Vin = 12 V; Vout = 30Volts.

In order to find the switching frequency, I used the formula:
Vo/Vi = Tsw/Toff = 1/(1-D).

Tsw I use comes from the ntype mosfet..IRFP60n. I got 55ns from the datasheet.

When I solve this for Tsw, I get Tsw = 1.38E-7.
From this, I get fsw = 1/Tsw = 7 Mhz.

I'm used to seeing boost converter frequencies in the kHz. I'm afraid this is too high and will result in losses. I also know there is the RHP zero issue but I'm not sure if that frequency will affect it and I'm not particularly sure how to calculate that.

I appreciate your help in advance.

Thanks
 

7MHz is way too high. You'll have too many losses. Like you said, there will be too many losses. Tsw is the period of the switching frequency. This is a value that you put in depending on your chosen frequency. So, if you decide to use 50kHz, Tsw becomes 1/50000 s. Toff is the time in this (1/50000)s that the switch stays off.

Substitute in your equation. You have:
30/12 = (1/50000)/Toff

Hope this helps.
Tahmid.
 

Hi,
First you need to understand the benefit of high frequency and its draw back,
But before that why to go for high frequency? There is good relation between L and C passive filter size at the output of your Boost reg and frequency. In fact high frequency reduce the responsibility of output Inductor and Capacitor and hence good regulation. So High frequency my reduce your Inductor and capacitor sizes but trade of with switching losses.
Thumb rule is we should use switching freq for regulators from few tens of kHz to 500-700 kHz.
Hope this may help you.
 

thanks. I've read a lot about boost converters in the last few days but none of the papers ever said how to choose the switching frequency. Do i just make an educated guess? I thought Toff was that of the mosfet; hence my reason for using 55ns which then made the frequency really high.

Also, how do i choose the resistor load? is that an educated guess also?

I apologize for these dumb questions but I've read a lot and still had this questions.

thanks
 

Hello,
In order to find the switching frequency, I used the formula:
Vo/Vi = Tsw/Toff = 1/(1-D).
Vo/Vi=1/(1-D) is a valid formula (if the converter is operating in continuous conduction mode), but that has nothing to do with Tsw. I'm not sure where you got that relationship from, but it's wrong.

Choosing a switching frequency is often not an exact science. It determines things like efficiency (higher frequency generally means lower efficiency), EMI (frequency affects the strength and frequency content of EMI), ripple (higher frequency means lower ripple), cost (higher frequency usually means higher cose), etc, all of which are competing requirements.

If you're just starting out with SMPS design I would stick to a frequency of 20KHz-100KHz.

---------- Post added at 09:14 ---------- Previous post was at 09:07 ----------

I thought Toff was that of the mosfet; hence my reason for using 55ns which then made the frequency really high.
Okay I can see your source of confusion. In a PWM controller, you want to drive the FET to conduct for a certain amount of time (often called Ton) and off a certain amount of time (Toff). This has nothing to do with the mosfet itself. However, the mosfet will have switching times, which define how long it takes a FET to change between on and off states. In the datasheet these are often referred to as Ton and Toff. But they're not the same as the Ton and Toff of the PWM waveform! It's just an unfortunate coincidence. If you are planning on making a fixed frequency converter (most boost converters are), then it makes sense to refer to the PWM waveform not by its Ton and Toff, but by its frequency and duty cycle. It still describes the waveform, but uses less confusing terminology.
Also, how do i choose the resistor load? is that an educated guess also?
Are you building the converter for a specific application and load, or just for fun/educational purposes? If the latter, I would try for a power level between 1-10W. So at 30V out, you'll want a load resistance of 90-900ohms.
 

Similar sounding formula signs doesn't necessarily mean the same.

You're apparently referring to td(off) Turn-Off Delay Time which has nothing to do with converter "on" or "off" times. To achieve low switching losses, the latter should be at least one or better two orders of magnitude larger than the transistor switching time (delays and rise-/fall time).
 

thanks. I've read a lot about boost converters in the last few days but none of the papers ever said how to choose the switching frequency. Do i just make an educated guess? I thought Toff was that of the mosfet; hence my reason for using 55ns which then made the frequency really high.

Also, how do i choose the resistor load? is that an educated guess also?

I apologize for these dumb questions but I've read a lot and still had this questions.

thanks

You choose the switching frequency depending on various parameters. You can start from say 50kHz. Then decide if this is an ok frequency considering losses, inductor and capacitor sizes, required ESR and value of capacitor, required inductance, etc. Then you decide if you can go higher and if this will yield benefits. If you see that going higher has no significant benefits and has higher losses, etc, go lower. I would say that 50-100kHz is a good range to choose from.

You determine the maximum power the boost converter is going to deliver. Then you use a load resistor that dissipates that much power. You can increase the resistance to see performance at lower power levels and you can decrease the resistance to see performance at higher powers, such as, (if you implemented it) the overload/overcurrent circuits, etc.

Hope this helps.
Tahmid.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top