MC34063 (MC33063) Timing Capacitor

Figure 1 of the data sheet shows that 2.6usec is right at the edge of capability of the MC34063A.  As a result, I'm thinking that your application may be pushing the part out of its comfort zone.  The MC34063A is an older BJT part.  It is inexpensive and has been used in countless applications, but needs to operate within limits to be successful.

What are your requirements?  Perhaps a newer part might give you better performance and predictability?

Thanks for the reply.

Cost is a major factor in my design, so that is why I picked the MC34063.  The application is in a buck configuration from 25-38V to 5V out @ 40mA - 400mA Iout.  I built the design with a 120pF capacitor and measured the Tchg of the timing capacitor to be about 6us -- the Ton of the switch was about the same at 5.8us.  This caused it to operate in discontinuous mode at the lower output current which, and this is ultimately why there is an issue, caused audible noise.  I've currently designed it for a 680uH inductor, which should have given me a max ripple of 72mA.  With a Ton of about 6us, I'd need to use almost 3mH of inductance to keep it from entering discontinuous mode at the highest input voltage and lowest output current.

Hi Jim,

I certainly understand the desire for keeping costs down.

As you've found, light load audio frequency is one of the "endearments" of the MC34063.  I had a car phone charger adapter that drove me nuts because of that, and the noise got into the AM band as well.  May I make a suggestion? 

A newer, higher frequency part could reduce the inductor size and cost significantly, offsetting the difference in IC cost.  For instance, a 400KHz to 500KHz operating frequency could reduce the inductor size to the 100uH range.  Digikey prices these in the $0.20 to $0.25 range versus $0.50 to $0.75 for a 680uH part with the same current rating.

The TPS40200 controller is $0.75 (TI web 1K price) versus $0.21 for the MC34063, and needs an external switch, which would cost around $0.10 to $0.20.  The inductor doesn't completely offset the price, but the total solution is much smaller.  Performance is better and more predictable.

http://focus.ti.com/docs/prod/folders/print/tps40200.html

I hope this helps.

I should have mentioned that size also isn't a factor, but I suppose that doesn't matter given the audio frequency issue.

At any rate, I've actually been pursuing the TPS5401 SWIFT buck switcher.  I've found though, that the current-mode control is much more complicated than the voltage-mode control and I'm having trouble with the compensation calculations for the TPS5401.

That's actually the problem.  When I use the equations in the TPS5401 application design guide, I get values that are unrealistic.  In fact, I can't even replicate the values that are achieved using the example parameters.

If you can post them along with your input parameters,I can take a look.  Perhaps you might have them in excel?

I do have them in excel.  See attached.

Also, I'm not sure this is the right thread for this anymore, but I'm not sure how to move it for the TPS5401 discussion.

"error 404 file not found"  You may need to repost

3162.TPS5401 Calculations.xlsx

I did some reading about voltage-mode and current-mode converters and I think that voltage-mode is better suited for my application.  David mentioned the TPS40200 with the external switch, but I am concerned about its operation at low output current (10-50mA).  The SwitherPro software doesn't even give the TPS40200 as an option when I give it a minimum current of 0.04A.

Also, when searching for inductors, I typically go to Coilcraft's website and for a given family of inductors, they are the same price for 56uH, 180uH and all the way up to 1000uH. So, I wasn't seeing a benefit to having a lower inductance (other than lower DCR).

I kind of jumped into this thread in the middle.  The best place to start is (of course) the beginning.  Can you tell me about your application and requirements?  Vin/Vout/Iout?  Then I can maybe point you in the right direction.

Should we move this thread to another since it doesn't apply to the original subject anymore?

As for the application, the input parameters are as follows:

Vin: 20-40V

Vout: 5V

Vripple: +/-5%

Iout: 30mA - 400mA

The only constraints I have are cost and keeping audible noise to a minimum. 

Also, were you able to open the TPS5401 calculations attachment?  I think there is something wrong with the equations in the data sheet.

yes why don't you start a new thread

Please mail the datasheet.

Non-Isolated DC/DC Forum