Help with step down 48V to 12V!!

[gumattos]

Well, I have no idea why you could not get the circuits I posted to work.

I tested the exact circuit below on a plug-in breadboard which is not optimum, and it worked the first time. I changed a few values to maximize the efficiency at 500 mA out and had it running for more than an hour. The only thing that slightly heated up was the 470 uH choke.

The Simulator predicted an efficiency of 87.65 % and I measured it at 86.1 %.
I suspect that the difference is being consumed in the choke. I will try a larger current choke later and see if it improves things.

The switching waveform on the drain of Q1 is nice with rise and fall times under 50 nS.

I will post actual scope plots a bit later.

From the beginning I never stopped considering the fact that I could be making some mistakes. If it works for it must work for me, I will buy brand new components and try again! Thank you very much E-Design for the effort you put in helping me, as soon as I can I will put results

 

[gumattos]

Well, I have no idea why you could not get the circuits I posted to work.

I tested the exact circuit below on a plug-in breadboard which is not optimum, and it worked the first time. I changed a few values to maximize the efficiency at 500 mA out and had it running for more than an hour. The only thing that slightly heated up was the 470 uH choke.

The Simulator predicted an efficiency of 87.65 % and I measured it at 86.1 %.
I suspect that the difference is being consumed in the choke. I will try a larger current choke later and see if it improves things.

The switching waveform on the drain of Q1 is nice with rise and fall times under 50 nS.

I will post actual scope plots a bit later.

Got it to work! Now I got the circuit to work, with low efficiency but working, 0,5A I get some bad eff. but I think it's only tweaking the values. One interesting thing I noticed, is with a 50mA load I got the circuit to not oscillate, do you think there is a way of making it better at small loads or the circuit by BradtheRad is better for small loads because of the opamp?

 

[E-design]

This type of circuit is somewhat load dependent for stable operation. This means that you need larger inductance values with low current. One thing you must make sure is that the choke you use is rated for the current required by the load.

If you use exactly the same values and circuit as I did, I see no reason why you have bad efficiency at 500 mA.

 

[gumattos]

This type of circuit is somewhat load dependent for stable operation. This means that you need larger inductance values with low current. One thing you must make sure is that the choke you use is rated for the current required by the load.

If you use exactly the same values and circuit as I did, I see no reason why you have bad efficiency at 500 mA.

The efficiency is bad because the circuit is no identical, I will get the components for identical circuit now, My only problem will be to find a 470uH inductor, but I have 2 220uH I will put in series

 

[E-design]

The thermal picture was taken after about an hour of operation. As you can see the choke is the only thing that heated up quite a bit.

I did find testing with BD139's from different manufacturers it does not always start up with the 18 k between the b-e of the BD140. So around 47k seems to be a happy median.

 

[BradtheRad]

OpAmp out:
View attachment 135173
As you can see, the output is no even close to the maximum ou minimum of the AmpOp, and since it's working as a comparator, this should be different.

The 741 output, at minimum is always a volt or two above its lower supply rail. You have it on a single supply. Therefore its output turns on an NPN continually. If you are still experimenting with this circuit, then you must turn off the NPN fully when it is supposed to be off.

Your resistor divider reduces the voltage a bit, but it's only a bit. Try putting a series diode or two at the op amp output. Try changing the resistor values. You need to make sure the NPN turns fully Off and fully On.

- - - Updated - - -

I was referring to your post #36 schematic:

View attachment 135160

 

[gumattos]

The 741 output, at minimum is always a volt or two above its lower supply rail. You have it on a single supply. Therefore its output turns on an NPN continually. If you are still experimenting with this circuit, then you must turn off the NPN fully when it is supposed to be off.

Your resistor divider reduces the voltage a bit, but it's only a bit. Try putting a series diode or two at the op amp output. Try changing the resistor values. You need to make sure the NPN turns fully Off and fully On.

- - - Updated - - -

I was referring to your post #36 schematic:

View attachment 135160

I replaced the UA741 with a LM393, I still need to put the series diodes?

- - - Updated - - -

The thermal picture was taken after about an hour of operation. As you can see the choke is the only thing that heated up quite a bit.

I did find testing with BD139's from different manufacturers it does not always start up with the 18 k between the b-e of the BD140. So around 47k seems to be a happy median.

Working Like a charm E-Design! At 0,5A 0,13A input 0,5A output, best efficiency until now. Do you think is there any way to make it work with low loads like 50mA? A high gain transistor maybe, changing the inductor, because a 48V to 12V converter like this one would help me a lot if it could handle lower loads, no problem if it loses efficiency with lower loads, if the current is low I can afford some bad efficiency, the only problem is that with the 50mA load it stops oscillating after some time. If it demands several changes on the circuit don't worry about it, you already helped me very much, BradTheRad's circuit seem to work on lower loads so I make both I guess.

 

[E-design]

I found a 470 uH choke with a higher current rating and now the efficiency at 500 mA went up to 89.2 %. So this should make you aware to check that the choke is DC rated for the load current.

 

[gumattos]

I found a 470 uH choke with a higher current rating and now the efficiency at 500 mA went up to 89.2 %. So this should make you aware to check that the choke is DC rated for the load current.

I am using a choke for 2.2A, more than enough I guess, but I am using two 220uH 2.2A chokes in series to get 440uH, that should be why I am getting a bit less efficiency. I tried to increase inductance to make it work with the 50mA load but no success so far.

 

[E-design]

Are your measurements to calculate the efficiency correct? Your figures indicate 96 %, which seems too good.

(12 * 0.5)/(48 * 0.13)*100 = 96.1 %

 

[BradtheRad]

I replaced the UA741 with a LM393, I still need to put the series diodes?

Since that is a comparator, it should turn off the NPN fully with no need for diodes to drop voltage.
However with its open collector, do you have a resistor network, etc., to turn on the NPN?

 

[gumattos]

Are your measurements to calculate the efficiency correct? Your figures indicate 96 %, which seems too good.

(12 * 0.5)/(48 * 0.13)*100 = 96.1 %

Sorry my bad, I got 11V on the output, probably because of the zener diodes (I ran out of 12V because I damaged the other in a test, so I bought 13V, now the regulation is 11V), and the local stores have no more of them. I ordered more on the internet, just waiting now. And I think my multimeter is a bit crazy, I will but a new one tomorrow.

- - - Updated - - -

Since that is a comparator, it should turn off the NPN fully with no need for diodes to drop voltage.
However with its open collector, do you have a resistor network, etc., to turn on the NPN?

I thought It was just like the picture below:


But the funny thing is that I forgot to connect the resistor, but it worked as well. I am now soldering the circuit (it was on a breadboard) with the resistor, please correct me if my connection is wrong.

Another thing I am thinking about is changing the 90k resistor for a 22nF capacitor, I think I got the idea of the capacitor feedback, will it work?


UPDATE: The resistor is 4.7k no 47k!!

 

[E-design]

The problem at light loads is that the choke will run dry and the regulator will operate in what is called "burst mode" as can be seen on the scope looking at gate and drain. This test was at 50 mA and the efficiency dropped to 56 %.

I found that it helps to put a shunt resistor of about 4.7 k over the 12 V zener, but it probably will affect the regulation.

 

[E-design]

I tried to increase inductance to make it work with the 50mA load but no success so far.

You will need something like 4.7 mH for stable operation at 50 mA

 

[gumattos]

You will need something like 4.7 mH for stable operation at 50 mA

I will try to get my hands in a 50mH indutor and make some tests then, let's see what happens.

 

[gumattos]

Some news, I am still searching for the higher inductance inductor, so I am trying to work on BradtheRad's design, I soldered it on a board, but something strange is happening. As you can see on the circuit, there is a resistor on the output of the comparator but the circuit works, with 67% efficiency. However this connection is ideal for the opAmp not the comparator, however when I replace the comparator for the opAmp, the efficiency goes to 24%. I tried capacitors for the hysteresis but no success so far.

 

[E-design]

Using a 4.7 mH choke at 50 mA will work well, but you will have to be sure that the choke can handle the larger currents without saturating.

The diagrams below show the response to a load changing between 50 MA and 550 mA.

----------------------------------------------

Your 393 needs a suitable output pull-up resistor to work.

 

[gumattos]

Using a 4.7 mH choke at 50 mA will work well, but you will have to be sure that the choke can handle the larger currents without saturating.

The diagrams below show the response to a load changing between 50 MA and 550 mA.

----------------------------------------------

Your 393 needs a suitable output pull-up resistor to work.

With the LM393, the problem is that when I put a 4,7k resistor between the output and the 24V the circuit goes to 24% efficiency, but without it gets 67%. Should I get the pull up from the 48V or from the 24V that is feeding the comparator? I used other resistor values but no success

 

[E-design]

You really need an oscilloscope to investigate what is going on. Without it, you are like the blind man on a busy street without a walking stick. You have to be able to see waveforms and measure switching frequency, etc.

The late **broken link removed** used to say something like: "It always makes me nervous when I find out that the customer I'm trying to help doesn't even have a scope".

 

[gumattos]

You really need an oscilloscope to investigate what is going on. Without it, you are like the blind man on a busy street without a walking stick. You have to be able to see waveforms and measure switching frequency, etc.

The late **broken link removed** used to say something like: "It always makes me nervous when I find out that the customer I'm trying to help doesn't even have a scope".

You can't be more correct E-Dseign, I will find a way to see what's going on. I will post the results here as soon as I can.