Author Topic: Reversible buck controller ?  (Read 1353 times)

0 Members and 1 Guest are viewing this topic.

Offline f4eruTopic starter

  • Super Contributor
  • ***
  • Posts: 1093
  • Country: 00
    • Chargehanger
Reversible buck controller ?
« on: July 04, 2022, 01:18:30 pm »
Hello,

I'm looking for a "reversible" buck controller, with or without integrated MOS, with sync rectifier, and 60V input max or higher.
Any ideas of a chip to use ?

I suspect most current-control loop based buck chips to not tolerate negative output currents, some specifically short the lower MOS when an output overvoltage appears.
The only chip I found which specifically allows reversing the current flow is the ISL6526 but it is too low voltage for the application
https://www.renesas.com/us/en/document/dst/isl6526-isl6526a-datasheet

Quote
The ISL6526, ISL6526A incorporate a MOSFET shoot-through
protection method which allows a converter to sink current as
well as source current. Care should be exercised when
designing a converter with the ISL6526, ISL6526A when it is
known that the converter may sink current.
When the converter is sinking current, it is behaving as a
boost converter that is regulating its input voltage. This
means that the converter is boosting current into the input
rail of the regulator. If there is nowhere for this current to go,
(such as to other distributed loads on the rail or through a
voltage limiting protection device), the capacitance on this
rail will absorb the current.


Online NiHaoMike

  • Super Contributor
  • ***
  • Posts: 9014
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: Reversible buck controller ?
« Reply #1 on: July 04, 2022, 02:19:38 pm »
LTC3871?
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 8172
  • Country: fi
Re: Reversible buck controller ?
« Reply #2 on: July 04, 2022, 04:16:18 pm »
I did this nearly a decade ago,

first iteration was a hysteretic peak current controller, current sense amplifier fed to two comparators (positive and negative overcurrent) and some logic gate thing to bootstrap gate driver. Worked acceptably. Some Cortex-M0 MCU controlled the two analog current setpoints, and measured current more accurately, finetuning the setpoints to give correct average current, and implement a very slow CV mode, too.

Second iteration was with STM32F334. This works nicely, CV mode is also faster.

But LTC3871 seems like a nice part. Did it exist back then? I think I searched the whole market gazillion of times and all I could find was some super over expensive and complicated ADI or LTC chipset requiring two ICs for the job. So I did it from scratch.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16614
  • Country: us
  • DavidH
Re: Reversible buck controller ?
« Reply #3 on: July 04, 2022, 06:16:20 pm »
There are 4-quadrant converters like the LT8714.

https://www.analog.com/en/analog-dialogue/raqs/raq-issue-183.html
 

Offline f4eruTopic starter

  • Super Contributor
  • ***
  • Posts: 1093
  • Country: 00
    • Chargehanger
Re: Reversible buck controller ?
« Reply #4 on: July 04, 2022, 07:47:42 pm »
Thanks,
interesting options, but those are quite expensive.
The DIY is also an option, but a lot of dev hours needed for SW, and additionnal HW complexity (2 LDO, one mosdriver, current sensing...)

I'll be investigating the behavior of some non-specified chips to see if I get one that could fulfill the reverse operation.
Candidates :
- MAX17504 2,5$ with transistors (in PWM mode?)
- ltc7891 in forced continuous mode
- ISL8117A perhaps with a bias on current sensing to allow for negative current
....

Offline Faringdon

  • Super Contributor
  • ***
  • Posts: 1990
  • Country: gb
Re: Reversible buck controller ?
« Reply #5 on: July 08, 2022, 08:31:12 am »
Good way is to just set up an H bridge, with inductor "in the middle", as normal.
Then pulse either set of fets depending on which way you want to go.
Do it synchronously or not.
Have analog switchs etc to switch the control , output dividers, current senses, and gate drive signals over etc.
I once looked into this for a Peltier current supplier..because the current had to go both ways.

So just use any standard controller, and have the "changeover switch" situation to switch from one way to the other.
« Last Edit: July 08, 2022, 08:32:45 am by Faringdon »
'Perfection' is the enemy of 'perfectly satisfactory'
 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 8172
  • Country: fi
Re: Reversible buck controller ?
« Reply #6 on: July 08, 2022, 10:33:57 am »
If the sides have a permanent voltage relationship, for example V1 > V2 always holds, then the two-switch topology is simpler and H bridge offers no advantages.

Two-switch configuration is equal to synchronous buck and synchronous boost, depending on the direction you look it from. With a H bridge, you can arbitrarily choose which side bucks and which boosts, so both V1 > V2 and V1 < V2 become possible.
 

Offline langwadt

  • Super Contributor
  • ***
  • Posts: 4422
  • Country: dk
Re: Reversible buck controller ?
« Reply #7 on: July 08, 2022, 11:31:59 am »
If the sides have a permanent voltage relationship, for example V1 > V2 always holds, then the two-switch topology is simpler and H bridge offers no advantages.

Two-switch configuration is equal to synchronous buck and synchronous boost, depending on the direction you look it from. With a H bridge, you can arbitrarily choose which side bucks and which boosts, so both V1 > V2 and V1 < V2 become possible.


like, https://www.maximintegrated.com/en/products/power/power-management-ics/high-performance-pmics/MAX1968.html
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16614
  • Country: us
  • DavidH
Re: Reversible buck controller ?
« Reply #8 on: July 08, 2022, 07:30:50 pm »
If the sides have a permanent voltage relationship, for example V1 > V2 always holds, then the two-switch topology is simpler and H bridge offers no advantages.

Two-switch configuration is equal to synchronous buck and synchronous boost, depending on the direction you look it from. With a H bridge, you can arbitrarily choose which side bucks and which boosts, so both V1 > V2 and V1 < V2 become possible.

If I understood you correctly, the advantage of an h-bridge is that continuous conduction mode can be maintained at all current levels, just like with any synchronous switch converter.
 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 8172
  • Country: fi
Re: Reversible buck controller ?
« Reply #9 on: July 09, 2022, 06:20:31 am »
If the sides have a permanent voltage relationship, for example V1 > V2 always holds, then the two-switch topology is simpler and H bridge offers no advantages.

Two-switch configuration is equal to synchronous buck and synchronous boost, depending on the direction you look it from. With a H bridge, you can arbitrarily choose which side bucks and which boosts, so both V1 > V2 and V1 < V2 become possible.

If I understood you correctly, the advantage of an h-bridge is that continuous conduction mode can be maintained at all current levels, just like with any synchronous switch converter.

No, continuous conduction can be maintained with synchronous half-bridge, too. The key to forced CCM is use two active switches instead a switch + diode, this is enough.

H bridge allows these configurations:
V1 > V2,  current flow 1 -> 2
V1 > V2,  current flow 1 <- 2
V1 < V2,  current flow 1 -> 2
V1 < V2,  current flow 1 <- 2

Half bridge only allows the first two (or latter two, depending on which way you define the labels).
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf