Can a boost regulator be connected in series to a buck regualtor?

[TWalker]

Hi, I'm having trouble finding a simple answer.
Can a boost regulator be connected in series to a buck regulator?
Reason being, I have a small 12v dc solar/battery/regulator supply. Discharge/recharge means the voltage shifts from 11v - 16v.
I wish to power a 14v 2.3amp led monitor, so neither a buck or boost alone will work.
I need to control both the Amp & voltage output.
I can buy both buck & boost modules with adjustable Amp output (& adjustable v output).
The combined buck/boost module has only an adjustable voltage & I have already burnt out a resister connecting it.
So any thoughts appreciated, thanks

 

[brushhead]

Hi,
You can get a topology known as buck/boost that will perform both functions. Have a look at this


https://en.wikipedia.org/wiki/Buck–boost_converter

At a 50/50 duty, the output volts equal the input volts. Any deviation will buck or boost the output.

 

[TWalker]

Hi,
You can get a topology known as buck/boost that will perform both functions. Have a look at this

thanks for you reply, Not sure if I've misunderstand you answer, I have read the wiki link earlier, but I don't think it addressed my question.
I have tried this "The combined buck/boost module has only an adjustable voltage & I have already burnt out a resister connecting it. "
I want to use what I have locally available "I need to control both the Amp & voltage output."
My understanding is Very basic, just trying to avoid damaging any more modules or my monitor. Not sure if things like A common ground might be an issue.
heres the specks;

Heavy Duty Boost Converter Constant Current 250W
DC DC boost converter Constant Current Mobile Power supply 250W 10A LED Driver

Specification:
Input Voltage: 8.5V-48V (48V maximum)
Input Current: 10A (max) if exceeds 8A please include heat dissipation
Quiescent current: 10mA
Output Voltage: 10-50V (adjustable)
Output Current: 10A (max) if exceeds 6A please include heat dissipation
Output Power: Approx. 250W
Operating Temperature: -40 to 85 degrees
Operating frequency: 150kHz
Dimension: 36 (L) x 70 (W) x 13 (H) mm
Weight: 50g

This Boost converter has overcurrent protection (input exceeds 13A) and reverse polarity protection capability.

Constant Current and Voltage Module 5A Adjustable
This is a non-isolated step-down, non-synchronous DC DC switch mode converter module.

This module uses high-quality and high efficiency ICs, in conjunction with the original TI op amp to control the output current to produce very stable output. The module also uses low ESR Sanyo original filter capacitor which reduces the ripple voltage to as low as 50 mV.

Applications:
-Normal blood pressure modules
-Battery Charger
- High-power LED constant current driver module
- Any applications that requires constant current constant voltage


Specification:
Input voltage: 5V - 32V
Output voltage: 0.8V -30V
Output Current: 5A maximum (Adjustable)
Efficiency: up to 95%
Switching frequency: 300KHz
Output ripple: 50mV (maximum)
Load Regulation: ± 0.5%
Voltage Regulation: ± 2.5%
Operating temperature: -40 ? to + 85 ?
Size: 51mm (L) x 26.3mm (W) x 14mm (H)
Overcurrent protection capability :Yes

Note: If more than 3A or 35W is used, please include fan or heat sink for heat dissipation purpos

 

[KlausST]

Hi,

Buck/boost is definitely an option.
Besides this SEPIC may be used (where the up/down feature is made with a series capacitor)

*****
Your specifications read like the cheap modules sold at ebay...

First you need a boost converter to generate a constant high voltage. This voltage needs to be higher than the max battery voltage and it needs to be higher than the max LED voltage plus some headroom.

From this voltage it is possible to run the constant current buck converter to drive the LEDs.

Theoretically possible. But since both regulators run independently with each on its own frequency, this may cause problems when both regulation loops influence each other. Consider filters between the modules, maybe even shielding boxes.

The buck stage needs to act like a constant current source. To achieve this you may modify typical constant voltage circuits by adding a current sensing shunt and an opamp to amplify it's voltage, this is the feedback to the regulator.
I recommend to use simulation like LTspice to check stability of the buck stage.

Klaus

 

[BradtheRad]

"I need to control both the Amp & voltage output."

Do specs for your LED monitor demand that it be supplied with only 14V?

Perhaps it can work with a supply of 15 or 16V? That is the upper limit you state for your supply.

Not sure if things like A common ground might be an issue.

This requires careful thought about the interaction between your power supplies.

One unit containing one supply can interact safely through a 2-conductor cable with another unit.

However is there a chance that power supplies in two different units, might interact by an unseen connection? Will a power supply in one unit, power circuitry in another unit which also contains a supply? (The concepts come across better by making block diagrams.)

Do any units have a connection to AC mains? This can give trouble.

It would be efficient, and simpler, if you could have one buck-boost converter. You can turn a boost type into a buck-boost, by disconnecting the output stage from ground, and connecting it to supply +. That's in its simplest form, although it may or may not be possible with your particular unit.

 

[brushhead]

I agree that running two power supplies would be asking for trouble. Have you ever heard of Middlebrook effect?

 

[treez]

this one can do the job for you...it can buck up or boost down, and it decides for itself which it has to do
https://www.linear.com/product/LT8705

 

[D.A.(Tony)Stewart]

Hi, I'm having trouble finding a simple answer.
Can a boost regulator be connected in series to a buck regulator?
Reason being, I have a small 12v dc solar/battery/regulator supply. Discharge/recharge means the voltage shifts from 11v - 16v.
I wish to power a 14v 2.3amp led monitor, so neither a buck or boost alone will work.
I need to control both the Amp & voltage output.
I can buy both buck & boost modules with adjustable Amp output (& adjustable v output).
The combined buck/boost module has only an adjustable voltage & I have already burnt out a resister connecting it.
So any thoughts appreciated, thanks

Yes you can BUT WILL get exception noise if the BOOST INPUT does not have really low ESR.

I once did this from 9~12 to 5V for a FPGA and low power circuits then boosted to 15V for AMLCD bias and video amp. The result was the boost choke was audible like running water until I put a large low ESR cap between converters. I later discovered this was a regulation instability called CHAOS. ( random noise) and that was only 100mA out at 15V.

- - - Updated - - -

create a free login to www.ti.com

https://webench.ti.com/webench5/pow...&O1I=2.3&O1V=14&op_TA=30&application=SWITCHER

 

[TWalker]

Do specs for your LED monitor demand that it be supplied with only 14V?

Perhaps it can work with a supply of 15 or 16V? That is the upper limit you state for your supply.

Do any units have a connection to AC mains? This can give trouble.
No its all solar/battery etc.

It would be efficient, and simpler, if you could have one buck-boost converter. You can turn a boost type into a buck-boost, by disconnecting the output stage from ground, and connecting it to supply +. That's in its simplest form, although it may or may not be possible with your particular unit.

Not sure if I can visualise that. I have to give it more thought.

Thanks for everyone's input.
LTspice is an awesome tool, but in this case, it would take more time than I have to verify the circuits.
The led monitor can be used with 12v to i4v, but searches on it have shown that it doesn't like too much variation & its life would be short. It wasn't a big investment, & at the end of the day its fun playing with this stuff. Ultimatumly I don't mind sacrificing time & a little money, if in the end I can gain confidence & have an efficient system. My error was not purchasing a new led monitor which generally run on 19v, although the transformers tend not to be separate & outside the monitor, some dismantling is needed.
I have be doing more research on it, still not confident it will work, but sacrifices will be made. Assuming they both are compatible, then the hardest thing may be getting the correct volt & amp output, as the values that I set (A&V) are maximum values the module will allow, eg I can get either the correct A or V, but may still end up with either a low A or V, I can have 1 but not necessary both. In the monitors case the correct amp may be ok, but I may end up with only 12v output. Fiddly tweaking all the values.

- - - Updated - - -

Yes you can BUT WILL get exception noise if the BOOST INPUT does not have really low ESR.

yeah that's also my concern, I was intending to boost 1st then buck - which has a Output ripple: 50mV (maximum), I was hoping that might clean it up, & I'm not sure if adding an additional capacitor & coil on the output will help.
I'll check out the info everyone's provided, thanks again, & then have a play & see how it goes :shock:

 

[TWalker]

Boost converter settings (series, 1st boost, then buck). Can anyone direct me to a formula to achieve the ideal input voltage & Amps into the buck converter to achieve 14volt, 2.14A output? (Efficiency: up to 95%)

 

[BradtheRad]

Although this idea is inefficient, did you rule out the idea of a boost converter (to 16V), followed by a resistive drop regulator (to 14V)?

Your monitor uses 14 x 2.3, or 32.2W.

You waste 1/8 of that amount, or 4W, which is not a severe amount, if you design proper heatsinking.

 

[treez]

Boost converter settings (series, 1st boost, then buck). Can anyone direct me to a formula to achieve the ideal input voltage & Amps into the buck converter to achieve 14volt, 2.14A output? (Efficiency: up to 95%)

TYPICALLY YOUR BUCK MODULE WILL NEED SOME 'HEADROOM' (sorry about caps)....so you need to be enough above 14 volt so as to account for the diode drop and the limitation of buck duty cycle to probably about 0.7....so make your input voltage = 14.7/0.7 = 21V or above.

Or if you wanted to be a bit canny you could make the input voltage to the buck so high that it always acted with D<0.5....... so go 14.7/0.5v input...its up to you, and which ever one you buy, you just have to check the voltage ratings to see if its possible,.

 

[KlausST]

Hi,

LTspice is an awesome tool, but in this case, it would take more time than I have to verify the circuits.

You are posting her for a couple of days now... be sure the simulation takes not that much time.

I'm not sure if adding an additional capacitor & coil on the output will help.

It will help.

Can anyone direct me to a formula to achieve the ideal input voltage & Amps into the buck converter to achieve 14volt, 2.14A output? (Efficiency: up to 95%)

Reading datasheets of the buck will give you the information. Also the simulation.

Klaus

 

[TWalker]

Using a resistive drop regulator to drop the v by 2v - 3v is diffidently an option, (I'd prefer more efficiency). I'm not clear what a "resistive drop regulator" is. Dropping a resistor in series on the output. Its a very simple method, I can control the A so just remove a few v, & its done.

Boost Output Voltage: 10-50V, so no problem with 21V or above ie (~29v). I will probably start there. thanks

 

[TWalker]

Of course your right Kluasst, I should get more familiar with LTspice.

 

[treez]

check out the LTMxxxx range on ltspice, they are SMPS's on a chip which may do exactly what you want.

 

[KlausST]

Hi,

Many manufacturers have interactive selection ti on guides for DC/DC converter ICs.

I often use the linear.com. (Although I can't say if other manufacturers are worse or better)
Just give in your voltage and current parameters and get a list of devices that meet your specifications.

I'd go for a SEPIC solution.

Klaus