Self-oscillating circuit for 5V power supply

[Miguel Lopez]

I'm trying to make a circuit using the components of a CFL (HV transistors, ferrite core, HV diodes, HV capacitors, etc) in order to obtain a 5V source to supply a TTL-made timer for a microwave oven. All circuit I have checked use two ferrite cores. I think it is posible to do it using just one core.

I need some help with the cicuit configuration (for the self oscilating circuit) and how many turns (approx.) should I coil in the ferrite core. The 5V circuit will demand 500mA.

I want to rectify the 110 V AC line supply and using one of the capacitor of the CFL, obtain 150 V CD (non-regulated), and then the self-oscilating circuit with one independent secondary to obtain 5V using a single diode and a small capacitor. I've read that 40 kHz oscilation would be suitable for this purpose.

It is now working with an iron core transformer and a 7805 (TO-3), but I want to change to ferrite core because the iron transformer and the 7805 get hot when they work for some time.

Thanks in advance.

Miguel

PS: If there is a thread about this I apologise. Please redirect me to that thread.

[G8HQP Dave]

This sounds dangerous, being in effect a DIY 'live chassis' circuit.

[Miguel Lopez]

No. My idea is to make an independent coil for the 5V circuit. This coil will be electrically isolated from the 150V circuit. The 150V circuit won't be ellectrically connected to anything but the line supply.

This is my idea (see attachment):

[kalee20]

Designing a switch-mode off-line power supply is not a 5-minute job. With the constraints of using the bits from an existing unit (ie not allowing for blowups with no replacements allowed) it's almost impossible.

If you want a 5V 500mA supply, can you not go to a car boot sale (do they have them in Cuba)? Me, i'd use a small mains transformer, rectifier, capacitor, 7805. Yes, it may get warm. But, no RFI, easier to make, msafer, more reliable, etc wins every time!

[kalee20]

My post crossed with the OP's...

Straight away, looking at the circuit, it won't work well.

The transformer phasing is wrong - it's operating in the forward mode, with no output choke (so indeterminate peak current in the switching transistor) and there's no energy recovery winding. What happens to the energy stored in the core - or is it intended to resonate with the capacitor across the transistor? Dodgy...

If you reverse the secondary wires, and rectify during the flyback period, most of the above is fixed. But then you'll have to either clamp the output voltage by a power Zener diode, to stop it rising if the load is removed, or find some way of controlling the primary-side to regulate it.

I'd stop now!

[Miguel Lopez]

Originally Posted by Kalee20

Designing a switch-mode off-line power supply is not a 5-minute job.

I know. That's why ask here for help.

Originally Posted by Kalee20

can you not go to a car boot sale (do they have them in Cuba)?

NEGATIVE. WE DON'T. (WITH CAPITAL LETTERS)

We do what we can, with whatever thing we have at hand.

I don't have a transformer suitable for that purpose. It really get hot, and I need to change it. That's why I thought in the CFL. CFL are availables at assequible prices.

Originally Posted by Kalee20

If you reverse the secondary wires, and rectify during the flyback period, most of the above is fixed.

Could you draw a diagram explaining those changes?

Originally Posted by Kalee20

I'd stop now!

I won't. That would mean no microwave oven in a few time. I will continue researching.

Thanks anyway.

[glowinganode]

Hi Miguel,
I admire your determination, but I'd start with something that was designed as a low voltage dc power supply.
A mobile phone charger could be a good idea, one for charging through a USB port would be ideal as it should be rated at 5Vdc.
Rob.

[mrmagnetophon]

I'd get a cell phone charger, those have to be readily available and cheap to you?

[Ed_Dinning]

Hi Miguel, if you are looking at simple SMPS designs then look at the applications on the Power Integrations website. These have plenty of design details as well as somer information on transformers. They normally use an integrated controller and mosfet in a TO220 package.

Ed

[Miguel Lopez]

Thanks to glowinganode and mrmagnetophon

When I read your post I was with one of those chargers in my hands. I think that could be a good idea. I can reverse-engineer this charger and then I can change its parts by the components of a CFL, in order to obtain the milliamps that I require. It could take a lot of "trial and error" (making emphasis in error), but I think it could work.

Sorry Ed, that would be nice but those integrated controllers are not available to me. Thanks for your cooperation.

[mrmagnetophon]

How many milliamps are you trying to achieve?
The cell phone charger transformer should be good for up to 300mA and with a few capacitors and maybe a new bridge rectifier built from cfl parts should be able to handle up to 500mA for non-constant usage.

-Chris

[Miguel Lopez]

I need 500 mA for non-constant usage. The cell phone charger I've seen in Havana are only up to 100 mA.

[Miguel Lopez]

Looking for info in the web I found this diagram. I think I can mount it.

http://www.seekic.com/uploadfile/ic-...3302203978.gif

Now I need to know how many turns must have each coil. I will measure the section area, and I hope all of you give some help in this matter.

[davegsm82]

Its surprising that your cell phone chargers are only 100mA, generally anything within the last 5-6 years will be at least 500mA to conform with the USB standard.

I guess things are very different in Cuba, but over here most people are falling over old phone chargers as they are everywhere.

Can you get in-car (automobile) Cell phone chargers cheaply? they usually provide 5V at up to 500mA to 1000mA, and they will run from any voltage from 7 up to 24v, you could use any old transformer/charger with one of these.

Dave.

[Miguel Lopez]

Hi Dave.

Friend, I can assure you that you would be surprised about the things that happen in Cuba. They would be surrealistic for you. Don't try to understand them or your brain would burn out. The only thing I can tell you about that, is .....yes, there are those chargers which can deliver 500 to 1000mA, but they are out of my reach. Economic reasons.

I have dismounted the CFL circuit I had, and then I will design the PCB to begin my "trial and error process". Anyway, I'll take a break during these days in order to spend some time with my family. Not everything has to be work and hobby.

[Miguel Lopez]

I found a description of how that circuit operates. It has several translation errors, but it's useful anyway.

For the market everywhere in the mobile phone charger,universal charge is growing rapidly, but are not very high quality, often a problem, throw a pity, it will teach you the recipe for analyzing the principles of cell phone chargers, hope that they can repair bring some help. Analysis of a power supply, often started from the input. 220V AC input, end of 4007 after a half-wave rectifier, the other end through a resistor 10 in Europe, by the 10uF capacitor filter. The resistance used for 10 protection in Europe, if followed as a result of over-current fault occurs, then the resistance will be blown, so to avoid a bigger failure. Right 4007,4700 pF capacitor, 82KΩ resistor to form a high-pressure absorption circuit, when the switch off when the 13003 is responsible for absorption of the induction coil voltage, thereby preventing the pressure added to the switch 13003 which led to breakdown. 13003 for the switch (complete name should be MJE13003), voltage 400V, the maximum collector current of 1.5A, the maximum collector power dissipation of 14W, is used to control the primary winding and power between the pass, off.

When the primary winding non-stop off, they will change in the switching transformer in the formation of the magnetic field, thereby inducing a voltage in the secondary winding. Because the figure does not indicate the end of winding of the same name, it can not be seen to be a forward or flyback. However, the structure of this circuit point of view, can be inferred that this power should be the flyback's. 510KΩ resistor to start the left side to switch to provide start with the base current. 13003 10Ω resistance below the current sampling resistor, the current into a voltage by sampling after (its value is 10 * I), this voltage by the diode 4148, the increase to the base of the transistor C945. When the sampling voltage is greater than approximately 1.4V, the switch current is greater than 0.14A, the transistor turn-C945, which will switch 13003 of the base voltage down to the collector current decreases, so that limits the switch current to prevent the current burn too large (in fact this is a constant current structure, the maximum switch current limit 140mA or so). Bottom left of the winding transformer (sampling winding) of induced voltage by the rectifier diode rectifier 4148, 22uF capacitor filter to form a sample voltage. For the convenience, we take the C945-emitter transistor to one end. Then this sample is the negative voltage (-4V or so), and the higher output voltage, the sampling voltage is more negative. After sampling voltage 6.2V regulator diode, added to the switch 13003 of the base. Said earlier, when the output voltage is higher, then the sample voltage more negative, as negative to a certain degree, 6.2V voltage regulator diode is the breakdown, which will switch 13003 of the base potential of down, which will lead to switch disconnect switch or delayed conduction to control the energy input to the transformer, also control the output voltage increases, the regulator output to achieve the function. The 1KΩ below 2700pF capacitor in series with resistance, it is positive feedback branch, removed from the sampling winding induced voltage, added to the switch on the base to maintain oscillation.

Secondary winding on the right there is not much to say anymore, rectified by diode RF93, 220uF capacitors filter the output voltage of 6V. Diode RF93 not find the information, it is estimated that a fast response control, such as Schottky diodes, because the high frequency switching power supply, so the frequency of the diodes need to work. Here you can use common 1N5816, 1N5817, etc. instead of Schottky diodes. Similarly, because the reasons for the high frequency, high frequency switching transformer must use the transformer core is usually high-frequency ferrite core with high resistivity to reduce eddy current.

http://www.gzhambo.com/NewsView.asp?ID=109&SortID=22

Can you help me with the calculations of turns?