Author Topic: Boost-buck for pump  (Read 1224 times)

0 Members and 1 Guest are viewing this topic.

Offline andrew_cTopic starter

  • Regular Contributor
  • *
  • Posts: 58
  • Country: gb
Boost-buck for pump
« on: March 05, 2020, 10:29:45 am »
Morning all,

Would anyone happen to know of a suitable series of boost-buck converters which will take in 22-34VDC and output 24V @ 2.5A. Digi-Key is not being my friend on this one.

Best regards,
Andy Cooke
 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 20362
  • Country: gb
  • 0999
Re: Boost-buck for pump
« Reply #1 on: March 05, 2020, 10:49:49 am »
Beware, buck-boost can refer to two different topologies.

Inverting: the output polarity is opposite to the input. It requires two switches, normally a transistor and a diode.

Non-inverting: essentially a buck converter, combined with a boost converter, but sharing the same inductor. It requires four switches: normally two transistors and two diodes.

https://en.wikipedia.org/wiki/Buck%E2%80%93boost_converter

Which one do you want?
 

Offline andrew_cTopic starter

  • Regular Contributor
  • *
  • Posts: 58
  • Country: gb
Re: Boost-buck for pump
« Reply #2 on: March 05, 2020, 11:02:49 am »
Non-inverting would be preferred.

Thanks, Andy
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Boost-buck for pump
« Reply #3 on: March 05, 2020, 01:08:12 pm »
Then SEPIC or "flying inductor" it is. :) Should be modules roundabout that level, not sure about Chinese ready made PCBs, think LT/AD makes a few chips if you want total integration, else any regulator or controller plus support components will do.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Online NiHaoMike

  • Super Contributor
  • ***
  • Posts: 9321
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: Boost-buck for pump
« Reply #4 on: March 05, 2020, 01:36:59 pm »
Check if the pump will run at 22V, which it most likely will. Then all you need is a buck converter.
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.
 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 20362
  • Country: gb
  • 0999
Re: Boost-buck for pump
« Reply #5 on: March 05, 2020, 01:41:15 pm »
Then SEPIC or "flying inductor" it is. :) Should be modules roundabout that level, not sure about Chinese ready made PCBs, think LT/AD makes a few chips if you want total integration, else any regulator or controller plus support components will do.

Tim
He did say buck-boost. I wonder why? Perhaps it's because it only requires one inductor?

Yes, there are plenty of off the shelf options.

This one meets the spec' at 24V 2.5A
https://www.digikey.co.uk/product-detail/en/mean-well-usa-inc/RSD-60G-24/1866-4203-ND/7706259

Although it's a pump, which might require more than that to start properly and the 101W model is only a pound more, so it's probably worth it.
https://www.digikey.co.uk/product-detail/en/mean-well-usa-inc/SD-100B-24/1866-4404-ND/7706462

As far as controller ICs are concerned, there are appear to be many on Digi-Key. Perhaps the original poster was being to strict in their search criteria? I had a quick look and found the LM5175, which looks good. It has a high efficiency, thanks to synchronous rectification.

http://www.ti.com/lit/ds/symlink/lm5175-q1.pdf
« Last Edit: March 05, 2020, 01:53:53 pm by Zero999 »
 

Offline angrybird

  • Regular Contributor
  • *
  • !
  • Posts: 142
  • Country: pr
  • I have a particular fascination with birds.
Re: Boost-buck for pump
« Reply #6 on: March 05, 2020, 02:03:06 pm »
What application is this for?  Load transients?  Ripple?  Efficiency requirements?  Any limits on switching frequency?  You are looking for parts intended for industrial (that should get you around 60V silicon) or Automotive (more like 40V tolerance).  Due to 24V out requirement I think industrial might have the most choices.  I have used some more recent buck-boost controllers (and converters with integrated FETs as well) which I think perfectly meet your requirements but unfortunately they are NDA-only parts, I just checked and there is not much public info on them.  However this means there are surely public options out there, but first think about your requirements.  You can sacrifice efficiency for really excellent performance with a buck and boost in series, or if efficiency is a big concern you could consider a first stage boost which can go into 100% duty cycle mode when Vin > Vout.  There are quite a few options here and I'm sure people will recommend different architectures but buck and boost should have great availability even in parts with integrated MOSFET.
THE CAKE IS A LIE AND THESE NUTHATCH ARE WAY TOO DISTRACTING
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Boost-buck for pump
« Reply #7 on: March 05, 2020, 02:05:24 pm »
Buck-boost is fine if the motor is isolated (not common ground) though.  This is a nice expedient, when available; lots of LED drivers are made this way for example.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline andrew_cTopic starter

  • Regular Contributor
  • *
  • Posts: 58
  • Country: gb
Re: Boost-buck for pump
« Reply #8 on: March 05, 2020, 03:27:38 pm »
The pump will run at 22V quite happily, thanks for spotting that.

This is for a class IIb medical application. We are not able to accommodate the size of the industrial units - certainly something we would be looking to bake into our existing circuit designs. An IC with integrated FET would be ideal.

The mains supply we're using is a class II with 300mV ripple - load transients should be an issue. We would like to see +90% efficiency.

Would most buck converters switch 100% on when Vin is below the desired Vout?
 

Offline angrybird

  • Regular Contributor
  • *
  • !
  • Posts: 142
  • Country: pr
  • I have a particular fascination with birds.
Re: Boost-buck for pump
« Reply #9 on: March 05, 2020, 04:01:44 pm »
You can certainly get bucks that will operate at "100%" duty cycle.  Just keep in mind that you still have the resistance of your high side FET and there will still be ohmic loss through the part even if the switching losses are not of concern.

Are you thinking that you could operate with a buck only?

If you are working with a legitamate business, you should have relatively easy access to parts covered by NDA, if you end up going the buck-boost route there are some nice ones out there.

One thing I might add is that generally speaking, the boost is going to be quite noisy...  What EMI/EMC certification are you going for?
THE CAKE IS A LIE AND THESE NUTHATCH ARE WAY TOO DISTRACTING
 

Offline Scrts

  • Frequent Contributor
  • **
  • Posts: 800
  • Country: lt
Re: Boost-buck for pump
« Reply #10 on: March 05, 2020, 04:53:18 pm »
For that current, you will need buck-boost controller with external FETs. Have a look here: http://www.ti.com/power-management/non-isolated-dc-dc-switching-regulators/buck-boost-inverting/buck-boost-inverting-controllers-external-switch/products.html

I'd say your search might be referenced with automotive grade truck power supply, since trucks use 24V systems.
 

Offline angrybird

  • Regular Contributor
  • *
  • !
  • Posts: 142
  • Country: pr
  • I have a particular fascination with birds.
Re: Boost-buck for pump
« Reply #11 on: March 05, 2020, 05:00:54 pm »
If he goes buck only, he could use an integrated part.  Dropout operation will need to be analyzed, though.  Dropout specifications for SMPS are "murky" at times :D
(i.e. 99% duty cycle may not actually mean 99% duty cycle  :-DD)
THE CAKE IS A LIE AND THESE NUTHATCH ARE WAY TOO DISTRACTING
 

Offline andrew_cTopic starter

  • Regular Contributor
  • *
  • Posts: 58
  • Country: gb
Re: Boost-buck for pump
« Reply #12 on: March 06, 2020, 10:44:29 am »
Yes, I believe a buck-only option would now be most suitable. Our business is rather new to the electronics market, this would only be our second commercial venture.

We shall be looking to comply with BS EN 60601-1-2:2015; which we've successfully navigated in the past.

The pump manufacturer has highlighted that the specific series of pumps we intend to use (not in my hand yet) are rated at just 35W, so just 1.5A if this helps with any further suggestions.

To meet the demands of the pump before the batteries internal cut-off is activated, we would need to see a duty cycle >98%. I'm certainly happy to handle a few parts to find one that works.
 

Offline fcb

  • Super Contributor
  • ***
  • Posts: 2135
  • Country: gb
  • Test instrument designer/G1YWC
    • Electron Plus
Re: Boost-buck for pump
« Reply #13 on: March 06, 2020, 11:01:17 am »
Anything can be made to comply with EMI - at a cost.

SEPIC is good if you need to meet your original specification and just want to run a simple motor. You will need to consider stall/fault conditions. TI make some great free selection and design tools for this:

If need to control pump speed (sounds like you don't), then I would use a boost configation (if you need>=24V) with a PWM MOSFET.

Probably wait till you get the pump in-stock and then carefully characterize it's performance in your real world application.
https://electron.plus Power Analysers, VI Signature Testers, Voltage References, Picoammeters, Curve Tracers.
 

Offline angrybird

  • Regular Contributor
  • *
  • !
  • Posts: 142
  • Country: pr
  • I have a particular fascination with birds.
Re: Boost-buck for pump
« Reply #14 on: March 06, 2020, 01:44:53 pm »
Yes, I believe a buck-only option would now be most suitable. Our business is rather new to the electronics market, this would only be our second commercial venture.

We shall be looking to comply with BS EN 60601-1-2:2015; which we've successfully navigated in the past.

The pump manufacturer has highlighted that the specific series of pumps we intend to use (not in my hand yet) are rated at just 35W, so just 1.5A if this helps with any further suggestions.

To meet the demands of the pump before the batteries internal cut-off is activated, we would need to see a duty cycle >98%. I'm certainly happy to handle a few parts to find one that works.

Just be aware that maximum duty cycle is a specification that suffers "marketing" effects at many IC manufacturers.  98% may not truly be 98%, the first thing you need to do is dig deeper, look at min off time (if they even specify it), get an evaulation board and look at it on the bench under your operating conditions.  Depending on what type of high side drive they have, look at the BST capacitor and make sure you aren't running out of charge at high duty cycle.  Companies like TI play a lot of games with the specs on their more competitive parts  :D
THE CAKE IS A LIE AND THESE NUTHATCH ARE WAY TOO DISTRACTING
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf