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.

DC voltage conversion

Status
Not open for further replies.

vipsam10

New Member
Hi.
I know this may sound ridiculous....but i need the simplest circuit which will convert 48DC to 24VDC with common grounds with a load requiring 1.2A. I know that 1.2A is too high for simple circuits, and that switch mode power supllies are required. But how to control the MOSFET switchings of the buck converter? I do not have any other available power supplies...any suggestion is most welcomed.....thx
 
Try a Black regulator, but the orgional circuit will need tweaking to be able to provide 1.2A at 24V with 48V in.

2-transistor Black Regulator
 
Hi hero999 well I did a 1.5 to 2.5amp black regulator a few weeks back for a psu I was building with a nice old 40v transformer to 13.8v output and a cabinet I had been saving.

It didnt turn out too well. I used a discrete darlington for the switch (ie 2 PNP transistors) so I could use base turn off resistors to get it to switch off fast enough but efficiency was much lower than I expected it was struggling to break 75% efficiency. The high saturation of the darlington was killing the efficiency. Rather than mess about with Sziklai pair, or mosfet driving circuits etc I just canned it and bought an off the shelf 240v to 13.8v 12 amp SMPS float charge PSU for about $25 USD new.

It's too often the case these days you can buy PSU stuff so cheap it's barely worth the labour of making it even if the parts are "free". :(

I do plan on having a go later with a 1.5amp black regulator using a mosfet, since so many people seem to want that 48v-24v or 48-12v in 1-1.5amp range since there's not that many low-cost simple solutions in that range.
 
why not simply use a LM78S40 (78s40) in buck mode.. it provides upto 1.5A load current
and if more be needed the you can add a by-pass pnp.. its all in the data sheet
 
Or even an MC34063...less pins smaller package...Downside is it needs an external switcher for the higher current..:(
 
Thx Sir. I consulted the datasheet for the LM78s40, but the maximum input is 40V....can i use 48V? Thx



why not simply use a LM78S40 (78s40) in buck mode.. it provides upto 1.5A load current
and if more be needed the you can add a by-pass pnp.. its all in the data sheet
 
It didnt turn out too well. I used a discrete darlington for the switch (ie 2 PNP transistors) so I could use base turn off resistors to get it to switch off fast enough but efficiency was much lower than I expected it was struggling to break 75% efficiency. The high saturation of the darlington was killing the efficiency.
For 48V to 24V you're better off with that than a linear regulator which would only be 50% efficient.


I do plan on having a go later with a 1.5amp black regulator using a mosfet, since so many people seem to want that 48v-24v or 48-12v in 1-1.5amp range since there's not that many low-cost simple solutions in that range.
The Black regulator isn't designed for MOSFETs.

You need a higher power better transistor than the BC327, the ZTX751 will do the job.

Try building the circuit designed for 24V input but obviously increase the zener to 24V or the nearest value you can find and uprate Q1. You probably won't even need to change any of the resistor values as the higher voltage will provide more current to the transistor's base.
2-transistor Black Regulator
 
Actually, my supply should be floating....i forgot to mention this part. I have a -48V supply (+ve terminal earthed) and i need to get -24V from it....can i use the same buck topology with the diode reversed?
Do you really need a floating supply? If you have -48V in and need -24V out, then they likely can share the same ground point and the regulator doesn't have to be floating.

You could use the Black regulator but you need to replace the NPN with a PNP, the PNP with an NPN, and reverse all the diode and capacitor polarities.
 
Yes, i presume i can use the black regulator, but the main problem is the relatively hign DC input and the 1.2A out. Do you think i can use the black regulator? Will the circuit be able to provide me with a current of >1.2Amps? Thanks
 
Yes, i presume i can use the black regulator, but the main problem is the relatively hign DC input and the 1.2A out. Do you think i can use the black regulator? Will the circuit be able to provide me with a current of >1.2Amps? Thanks
Yes, you just need to use transistors with a high enough voltage rating and a large enough current rated output transistor. (See the paragraph on "High Voltage Use" in 2-transistor Black Regulator.)

In spite of HERO999's reservations, I don't see any problem with using a high current N-MOSFET (for your desired negative voltage input and output) for the output transistor Q1 in place of a bipolar to make it easier to drive. Just adjust the value of R1 and R2 to bias the gate at about 10V gate-to-source voltage when the Q2 (now PNP) transistor is on. Since R1 and R2 no longer have to provide significant current to a bipolar transistor base, they can be reasonably large (in the low kΩ region) to minimize resistor power loss.

You also need an inductor that can handle >1.2A without saturating.
 
In spite of HERO999's reservations, I don't see any problem with using a high current N-MOSFET (for your desired negative voltage input and output) for the output transistor Q1 in place of a bipolar to make it easier to drive. Just adjust the value of R1 and R2 to bias the gate at about 10V gate-to-source voltage when the Q2 (now PNP) transistor is on. Since R1 and R2 no longer have to provide significant current to a bipolar transistor base, they can be reasonably large (in the low kΩ region) to minimize resistor power loss.
The reason for my reservations is that MOSFETs have a huge gate capacitance which would need to be charged and discharged by R1 and R2. This would take time and would slow the switching time unacceptable level leading to large losses.

You also need an inductor that can handle >1.2A without saturating.
No, the inductor would have to be able to handle much more than 1.2A. With 48V in and 24V out the duty cycle will be 50% so the peak current will be 2.4A. I would recommend a 2.5A or even a 3A inductor.
 
No, the inductor would have to be able to handle much more than 1.2A. With 48V in and 24V out the duty cycle will be 50% so the peak current will be 2.4A. I would recommend a 2.5A or even a 3A inductor.

You would also have to account for current ripple, which would make it 3.36A, so look for a 3.5A inductor
 
...

The Black regulator isn't designed for MOSFETs.

...

Umm, yeah I know, I designed it. :)

To vipsam10; you can use the black regulator for negative supplies with no problems, just invert everything, ie zener, electro caps and change NPN to PNP etc.

...
No, the inductor would have to be able to handle much more than 1.2A. With 48V in and 24V out the duty cycle will be 50% so the peak current will be 2.4A. I would recommend a 2.5A or even a 3A inductor.
...

It's not that bad really, the average inductor current is 1.2A (is the output current) and with duty cycles near 50:50 (like the OP requires) the current ripple is quite reasonable as the circuit oscillates around the inductor characteristics it will find a "sweet spot" where it oscillates easily and the current ripple would not be more than about 25%, so a 1.5 amp inductor will be fine.

If vipsam10 really wants to use a black regulator he could use an NPN darlington, or one of the Zetex NPN devices with a >80v Vce or higher. I would probably go the darlington as the purchase of a specialty transistor probably puts the project in the same league as buying a specialty chip, like the one blueroom suggested.
 
Umm, yeah I know, I designed it. :)
Sorry but I find that hard to believe: if you did design it, I doubt you would have made some of the errors in your above post but feel free to prove me wrong.

It's not that bad really, the average inductor current is 1.2A (is the output current) and with duty cycles near 50:50 (like the OP requires) the current ripple is quite reasonable as the circuit oscillates around the inductor characteristics it will find a "sweet spot" where it oscillates easily and the current ripple would not be more than about 25%, so a 1.5 amp inductor will be fine.
It's not the average inductor current that's important, it's the peak current. If the current exceeds the inductor's absolute maximum rating it will saturate causing a huge drop in inductance and a short circuit.

If vipsam10 really wants to use a black regulator he could use an NPN darlington, or one of the Zetex NPN devices with a >80v Vce or higher. I would probably go the darlington as the purchase of a specialty transistor probably puts the project in the same league as buying a specialty chip, like the one blueroom suggested.
The TIP31A will probably do but it won't be as good as the Zetex. There are plenty of other high quality BJTs which will do the job and let's also not forget the possibility of using a couple of BC327s with emitter ballancing resistors.

If more gain is required then Sziklai pair it but as I said before, I think a Sziklaior Darlington is not necessary because with 48V in the base current will be double, meaning the output current will also double.

I will do some simulations on it using LTSpice.
 
It turns out i was wrong about the peak current.

This regulator works in continuous conduction mode.

When simulated it at 300mA, the peak current is 431mA, assuming this ratio doesn't change much for 1.2A the current will be 1.73A so use a 2A inductor.

EDIT:
I was simulating the 5V out version.

I've just simulated the 12V out version at 300mA and peaks at 470mA, still a 2A inductor should do.
 
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top