Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

A combination of ideas

Status
Not open for further replies.
Ok I looked at it again and I stand corrected, but the super caps are changed through the first buck, which as you say isn't very efficient (you're right, high voltage bucks aren't and are current limited, that one is 1A not 1.5A continuous).

Given my intention to use the PIC for the 2 minute shutoff and to gather stats for pushing to RS232 why not do the logic via the PIC?

Also note, your last buck will drain the supercaps to 4.3V which is below USB spec. This can cause many phones/powerbanks to lock up until the power is removed completely (I've experienced this before).

Running through the logic, if the supercaps began charging and "fought" with the 5V going to the device, the first buck would probably trigger it's PG causing the supercaps to discharge?
 
Also note, your last buck will drain the supercaps to 4.3V which is below USB spec. This can cause many phones/powerbanks to lock up until the power is removed completely (I've experienced this before).
Don't understand.
The second buck discharges the super cap. There are two resistors on the LT3971-EN pin that set when the buck shuts down. I thought I have it giving up at about 6 volts, (on the super cap).

For most of the testing I have a 1A load at the output of the first buck and a different 1A load at the output of the second buck. That works well.
When I tied the two "+5V" together; then I have a problem. The second buck picks up the load and caused the first buck to think that power is good. I think if I play around with "under voltage lock out" on the first buck the problem can be fixed. The first buck should stop working because of under voltage and keep PG2 low. (there is my error! I thought that "under voltage" will cause PG2=0 but not true)
Back to Plan "A"
My first try: set the first buck to output 5.1V and the second to output 4.9V. That works.
 
That was the other potential issue I'd thought about, bucks don't like fighting each other. I've been using LTSpice recently, if you want to send the asc file over I can run it and see how well it works.

I still believe a PIC is necessary. Imagine the supercaps are dead and the input power = output power. There'd be pulsating power to the 5V out. This is another issue I've had previously which can lock up USB devices. I tried resolving it a while back with caps and transistors shorting them out but it wasn't effective.

Those two resistors, don't they determine the voltage out not the PG signal? Normally a buck won't shut down if the voltage isn't met, it'll have a % (I read 0.91% for that chip) before the PG is triggered but it wont shut the chip down. This could be made to trigger the EN pin but then delay and hysteresis comes into play.

I'm a programmer not an electrical engineer. The idea that I can get the values into the PIC and work the logic there will always sound more attractive!
 
I separated the two +5V outputs again.
The input is 400hz and run 5 cycles and stops. This gives a chance to see power up and power down conditions.
 

Attachments

  • Charnley-3.asc
    9.1 KB · Views: 133
OK, I'll fire that up in a bit.

I've been looking into PIC 10 bit ADC, which at 5V indicates 10mv resolution. I wonder if it'll be enough to prevent the FET backflow (or if it'll make any difference).
 
So you have two Vout's which should be connected or switched to bypass the buck-boost-buck series efficiency drop, but then the first buck is ramping up it's voltage which fails to power many USB devices. It should be turned on at PG but it looks like you're using the first buck to do that to the third (second buck).

Also, the circuit doesn't work with a 6V SINE.
 
Also, the circuit doesn't work with a 6V SINE.
If you make the input diodes Schottkey you will have less loss.
Active diodes are very complicated!
To get 5V out you probably need 6V. I did not look up the drop out voltage of the first buck.
Resistors also set the drop out (enable) voltage on the part.
You did not want to pull more than 500mA from the source BUT to get 5V at 1A out you will need 6V at 1A. (11V in at 500mA) I think at low speed you can not get this much power from the dynamo.
---------------------------------------------
Another option is to make the first stage operate all the time. It will get what ever power is available. Maybe only to half supply the output. The other half will come from the super cap.
(boost/buck--super cap--buck--5Vout)
I don't where you can get a boost/buck that will work from 5V to 100V.
 
Here is a part that you can learn from.
20V max input.
100mA max output.
Many good ideas!
 

Attachments

  • DN483.pdf
    247.5 KB · Views: 114
High voltage schottkey's don't have the significant voltage drop difference over silicons, it's another reason for attempting to use active FETs. You're right on the 100v buck though, they don't exist in higher current output / high sync efficiency. I'm still intending to use a PIC to switch various stages on and off. No idea if it'll work but shall soon find out!
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top