Helpful ReplyHot!DC/DC Boost Converter

Page: 123 > Showing page 1 of 3
Author
naeem1234
Super Member
  • Total Posts : 449
  • Reward points : 0
  • Joined: 2015/02/19 06:39:28
  • Location: 0
  • Status: offline
2018/10/13 15:37:42 (permalink)
0

DC/DC Boost Converter

Hello everyone
 
I am looking for some help about how to implement a DC/DC boost converter with a dsPIC33? Your help is highly appreciated.
 
#1
Bob White
Super Member
  • Total Posts : 197
  • Reward points : 0
  • Joined: 2010/11/06 19:52:38
  • Location: Denver, Colorado
  • Status: offline
Re: DC/DC Boost Converter 2018/10/13 21:07:27 (permalink)
+1 (1)
Aside from coding issues, the boost converter can have one of several different control-to-output transfer functions depending on whether you operate the converter in continuous or discontinuous conduction mode and whether you use voltage mode control or peak current mode control.

Some more information would help.
 
Input voltage (or input voltage range)
Desired output voltage or output voltage range
Output current/power range
 
Do you know if you want operate in continuous conduction mode (causes a right-half-plane zero in the control to output transfer function that complicates the controller design and also has higher losses from the reverse recovery of the boost diode) or discontinuous conduction mode (has higher peak currents and high rms current losses)?
 
Do you want to operate using voltage mode control or peak-current-mode control?  Voltage mode control is easier to implement (duty cycle based pulse with modulation) but current mode control simplifies the control-to-output transfer function.
 
#2
naeem1234
Super Member
  • Total Posts : 449
  • Reward points : 0
  • Joined: 2015/02/19 06:39:28
  • Location: 0
  • Status: offline
Re: DC/DC Boost Converter 2018/10/13 21:34:23 (permalink)
0
At present I have following information known to me.
 
Input voltage (or input voltage range) = 48 V DC (4 x 12 V)
Desired output voltage or output voltage range = 400 V (DC Bus bar)
Output current/power range = Approx. 7kW
 
The hardware circuit is already there.. which has the high-voltage drivers as well as analog inputs etc. 
 
I am not clear about the Voltage mode OR Current mode. How can i find out which of these two is implemented in the hardware?
 
Same for the Continuous mode OR Discontinuous mode.
 
#3
PStechPaul
Super Member
  • Total Posts : 2094
  • Reward points : 0
  • Joined: 2006/06/27 16:11:32
  • Location: Cockeysville, MD, USA
  • Status: offline
Re: DC/DC Boost Converter 2018/10/13 23:28:56 (permalink)
+1 (1)
If you just want to boost 48VDC nominal to 400 VDC nominal for a VFD, you may want to use a full H-bridge driven by PWM from the PIC, and driving a ferrite transformer followed by a FWB rectifier. I have been working on a very similar project that will use four 12V SLA batteries through the DC-DC converter and get 250-350 VDC for a 240 VAC VFD motor drive. Here is one of many simulations:

 
I made this prototype, using a PIC16F1825:

 
Here is one of the output waveforms before the rectifiers:


 
#4
naeem1234
Super Member
  • Total Posts : 449
  • Reward points : 0
  • Joined: 2015/02/19 06:39:28
  • Location: 0
  • Status: offline
Re: DC/DC Boost Converter 2018/10/14 05:24:33 (permalink)
0
PStechPaul
If you just want to boost 48VDC nominal to 400 VDC nominal for a VFD, you may want to use a full H-bridge driven by PWM from the PIC, and driving a ferrite transformer followed by a FWB rectifier. 



Yes i exactly want it in that application.. in my case i have to generate 3-phase PWM from the dsPIC33. Can you describe how the code would do that?
#5
PStechPaul
Super Member
  • Total Posts : 2094
  • Reward points : 0
  • Joined: 2006/06/27 16:11:32
  • Location: Cockeysville, MD, USA
  • Status: offline
Re: DC/DC Boost Converter 2018/10/14 19:44:04 (permalink)

 
#6
naeem1234
Super Member
  • Total Posts : 449
  • Reward points : 0
  • Joined: 2015/02/19 06:39:28
  • Location: 0
  • Status: offline
Re: DC/DC Boost Converter 2018/10/14 21:04:03 (permalink)
0
Thanks for the VFD Links.
 
Are there any App Notes or Examples for 3-Phase DC-DC Boost Converters?
#7
PStechPaul
Super Member
  • Total Posts : 2094
  • Reward points : 0
  • Joined: 2006/06/27 16:11:32
  • Location: Cockeysville, MD, USA
  • Status: offline
Re: DC/DC Boost Converter 2018/10/15 16:04:53 (permalink)
+1 (1)
Three phase usually applies to the AC signal, and has no meaning for DC. There are multiphase switch mode power supplies, which use interleaved waveforms to reduce noise and increase power, especially for synchronous rectification, but that is usually needed only for low output voltage buck converters.
 
http://ww1.microchip.com/...en/AppNotes/01114A.pdf
 
http://www.onsemi.com/pub/Collateral/AND8045-D.PDF
 
http://ww1.microchip.com/...AppNotes/90003155A.pdf
 
Perhaps it would be best for you to post the full specs of what you are designing, and preliminary schematics and block diagrams. It sounds like you want to run a 10 HP AC induction motor on 4 lead-acid batteries, which is similar to what I am doing. I made a DC-DC converter that was powered by two small 12V batteries in series, and about 250 VDC output, which powered a VFD and a 2 HP induction motor, on a small tractor.
 
https://www.youtube.com/watch?v=y0qWY4bVnEA
https://www.youtube.com/w..?v=j5TyhdY-cHQ&t=1s
post edited by PStechPaul - 2018/10/15 16:10:20

 
#8
qhb
Superb Member
  • Total Posts : 7891
  • Reward points : 0
  • Joined: 2016/06/05 14:55:32
  • Location: One step ahead...
  • Status: offline
Re: DC/DC Boost Converter 2018/10/15 16:09:15 (permalink)
+1 (1)
naeem1234
3-Phase DC-DC Boost Converters?

This is an "oxy-moron", i.e. a contradiction in terms.
Either you don't understand some of the requirements yourself, or you are not telling us the full story.
 
#9
naeem1234
Super Member
  • Total Posts : 449
  • Reward points : 0
  • Joined: 2015/02/19 06:39:28
  • Location: 0
  • Status: offline
Re: DC/DC Boost Converter 2018/10/16 09:42:37 (permalink)
0
qhb
naeem1234
3-Phase DC-DC Boost Converters?

This is an "oxy-moron", i.e. a contradiction in terms.
Either you don't understand some of the requirements yourself, or you are not telling us the full story.
 



Yes you are right. I was not clear about the requirements earlier. Now I am.
 
I just have to generate 3-phase PWM's that will be fed to a three phase power transformer using 48 VDC MOSFET full bridge . The output of the transformer will be 3-phase pure sinusoidals at 380 V 60 Hz. 
 
Is there any Microchip Example code to generate 3-phase sinudoidal using dsPIC33?
What should be the switching frequency of the PWM.. my initial guess is 20 kHz.. will that be ok?
#10
qhb
Superb Member
  • Total Posts : 7891
  • Reward points : 0
  • Joined: 2016/06/05 14:55:32
  • Location: One step ahead...
  • Status: offline
Re: DC/DC Boost Converter 2018/10/16 12:20:54 (permalink)
+1 (1)
naeem1234
...
I just have to generate 3-phase PWM's that will be fed to a three phase power transformer using 48 VDC MOSFET full bridge . The output of the transformer will be 3-phase pure sinusoidals at 380 V 60 Hz. 

So this is not "DC-DC" at all.
You just confused everyone by throwing in that incorrect term.
 
 
#11
du00000001
Just Some Member
  • Total Posts : 2238
  • Reward points : 0
  • Joined: 2016/05/03 13:52:42
  • Location: Germany
  • Status: offline
Re: DC/DC Boost Converter 2018/10/16 13:08:11 (permalink)
+1 (1)
Obviously you're intending to implement an inverting targeting either an island supply or a grid connection.
You might find application notes and more looking (searching) for "PV inverter", "solar inverter" and alike.
 
Depending on your ouput transformer and its peripherals, you might start as low as 8 kHz. On the other end the limit might be around 25 kHz - limited by switching delays and switching losses of your MOSFETs.
An aside: this doesn't seem to be a "boost converter" as the conversion to the higher voltage level is achieved by a transformer, not by chokes.

PEBKAC / EBKAC / POBCAK / PICNIC (eventually see en.wikipedia.org)
#12
PStechPaul
Super Member
  • Total Posts : 2094
  • Reward points : 0
  • Joined: 2006/06/27 16:11:32
  • Location: Cockeysville, MD, USA
  • Status: offline
Re: DC/DC Boost Converter 2018/10/16 13:37:33 (permalink)
+1 (1)
So you can make a three phase H-bridge for 48 VDC using power MOSFETs rated at 100 volts to make about 32 VAC. This can be fed to a conventional 3 phase 60 Hz 7 kVA power transformer with input windings of 32 VAC and output of 380 VAC. Or use two 3.5 kVA transformers. However, the transformer(s) will be rather large and heavy, probably as much as 50-100 pounds.
 
If you use low voltage MOSFETs, instead of high voltage IGBTs, you can probably use a higher PWM frequency, up to about 100 kHz. But the MOSFETs will need to handle up to 150 amps, so you will need multiple devices in parallel. This will still be required for a DC-DC converter, but once you get the ~530 VDC the VFD will only need to handle about 10 amps for 7 kW.
 
It is unclear why you need to drive a 60 Hz power transformer. If all you need is 380 VAC three phase, for driving a motor or almost any other load, the VFD can do that directly, and the components will be much smaller and more efficient.
 
If you just want a one-off solution, you can use the front end of commercially available inverters that produce 250-350 VDC to get modified sine wave 220 VAC. You can use 24 VDC into each of two such inverters, and connect the DC outputs in series, to obtain the required ~500-700 VDC, and feed that to the DC link of a commonly available 7 kVA (10 HP) 480 VAC VFD.
 
https://www.aimscorp.net/24-Volt-Modified-Sine-Inverters/
 
https://www.aliexpress.com/item/MKM5000-241G-off-grid-modified-sine-wave-inverter-5000-watt-inverter-5000w-24-volt-dc-to/32798330912.html?spm=2114.search0104.3.224.23a72b7dk9Zezx&ws_ab_test=searchweb0_0,searchweb201602_4_10065_10068_10130_318_10547_319_10546_317_10548_10545_10696_450_10084_10083_10618_452_535_534_533_10307_532_204_10059_10884_10887_100031_320_10103_448_449,searchweb201603_60,ppcSwitch_0&algo_expid=a0396b56-296f-4c9f-92d6-eb706e0e1b5f-36&algo_pvid=a0396b56-296f-4c9f-92d6-eb706e0e1b5f&transAbTest=ae803_5
 
https://www.aliexpress.com/store/product/6000W-peak-3kVA-ren-sinus-inverter-12-volt-24-volt-48-volt-home-inverter-3000w-pure/1735160_32847266086.html?spm=2114.12010608.0.0.28a0627dENtYOL
 
https://www.ebay.com/itm/1000W-8000W-Solar-Power-Inverter-DC-12V-24V-to-AC-110V-220V-Sine-Wave-Converter/123153415972
 

 
#13
Bob White
Super Member
  • Total Posts : 197
  • Reward points : 0
  • Joined: 2010/11/06 19:52:38
  • Location: Denver, Colorado
  • Status: offline
Re: DC/DC Boost Converter 2018/10/16 20:51:19 (permalink)
+3 (3)
I take from the original post and discussion that the OP wants to make a 48 Vdc to 3 Phase, 380 V, 50 Hz output that may, or may not, be grid connected.  As I understood the original post, this converter would have two stages.  The first is a boost converter to make about 400 Vdc from the 48 Vdc.  Then there would be a three phase, transformerless inverter, to make the 380 Vac/50 Hz.  This approach is not uncommon but has real problems with efficiency.
 
With a boost factor of almost 10 (48 Vdc is really more like 42-44 Vdc under load and discharge) to 400 Vdc) a standard nonisolated boost converter would be running at about 90% duty cycle (Vout/Vin = 1/(1-D)).  With 7 kW of output power and assuming an unrealistic 90% efficiency, the dc input current would be about 175 A.  Perhaps with very high capacity, very heavy duty, high discharge rated marine or telco central office batteries one could provide that current for an hour or two, but not with a typical automotive battery.  Unless there is some other source of input power other than a battery I think the original poster has taken on an unrealistic task.
 
One can make an appropriate inductor for the boost converter but it would the size of a load of bread and could weigh 10-20 pounds.  The switching devices for the boost converter could be a MOSFET module but a more practical approach would be an IGBT module rated for at least 600 V and 300 A.
 
The inverter bridge could be either MOSFETs or IGBTs rated at least 600 V and 50 A.  There are many IGBT "six packs" available that would fulfill the need and could work with switching frequencies up to about 20 kHz.
 
The original poster has hinted that this is a grid connect application.  That brings a whole host of other problems.  One problem is converter stability with the required output filter.  The LC filter of the inverter combines with the inductance of the grid to make an LCL filter stage that creates real control problems.
 
Another problem is anti-islanding.  For the US and Europe where the grid is well regulated for both voltage and frequency there is enough challenge to detect when to disconnect (such as when the grid frequency drifts off target by 0.1 Hz or less).  In countries where the grid is less well regulated, this becomes a problem.  If the grid frequency drops from a nominal 50 Hz to 48 Hz, is that just a normal grid fluctuation or is it indicator that the local grid is overloaded and failing and that the inverter should disconnect?  If the inverter keeps powering the local grid when the main grid power has failed that can create a real safety problem for repair personnel.
In summary, this is a far from trivial project that requires substantial expertise in power electronics and power electronics control to succeed.  I have seen no real evidence that the original poster or any of his/her associates have the required expertise.
 
post edited by Bob White - 2018/10/17 14:08:23
#14
naeem1234
Super Member
  • Total Posts : 449
  • Reward points : 0
  • Joined: 2015/02/19 06:39:28
  • Location: 0
  • Status: offline
Re: DC/DC Boost Converter 2018/10/17 00:51:03 (permalink)
0
du00000001
Obviously you're intending to implement an inverting targeting either an island supply or a grid connection.
You might find application notes and more looking (searching) for "PV inverter", "solar inverter" and alike.
 



With time I am getting more and more details of my task in this project which was not very clear to me few days back. To be exact my task is just to create 3-phase 60 Hz Sinusoidal PWM output at 20 kHz switching frequency using a dsPIC33E MCU. I wonder what could be the best approach for this.. either using a SINE_TABLE[] array of pre-calculated values or doing run-time trigonometric calculations or maybe some other more elegant approach?
post edited by naeem1234 - 2018/10/17 00:57:54
#15
PStechPaul
Super Member
  • Total Posts : 2094
  • Reward points : 0
  • Joined: 2006/06/27 16:11:32
  • Location: Cockeysville, MD, USA
  • Status: offline
Re: DC/DC Boost Converter 2018/10/17 01:55:06 (permalink)
+1 (1)
I would suggest that you build a prototype using the dsPIC33E if that is a design requirement, and then use some of the application notes and reference designs to get a three phase output using perhaps just 12V or 48V into a low power resistive load. I would suggest a sine table for just 1/4 wave, or 90 degrees. The other quadrants can be derived by working back through the table for 90-180, and inverting the values for 180-360. Once you get this simple prototype working, you will have some of the experience required to proceed to the next steps.
 
I am actually planning to do something very similar, but it will be a simple VFD for a variable speed 3 phase motor control of about 2 HP. It will use batteries as this will be for a small tractor, and I will probably use a couple of 3000W peak, 1500W continuous, 24V to 220V inverters to get about 600 VDC for the VFD bus. I found some for only $32 each:
https://www.ebay.com/itm/3000-4000-5000-6000W-Car-Power-Inverter-Converter-DC-12V-24V-to-AC-110V-220-240V/123324152681?var=423722983675

 

 
#16
naeem1234
Super Member
  • Total Posts : 449
  • Reward points : 0
  • Joined: 2015/02/19 06:39:28
  • Location: 0
  • Status: offline
Re: DC/DC Boost Converter 2018/10/17 08:21:01 (permalink)
0
PStechPaul
I would suggest that you build a prototype using the dsPIC33E if that is a design requirement,...



Can you suggest some other MCU than a dsPIC33E for the purpose of generating 3-phase sinusoidal PWM at 60 Hz?
#17
naeem1234
Super Member
  • Total Posts : 449
  • Reward points : 0
  • Joined: 2015/02/19 06:39:28
  • Location: 0
  • Status: offline
Re: DC/DC Boost Converter 2018/10/18 09:16:48 (permalink)
0
PStechPaul
I would suggest a sine table for just 1/4 wave, or 90 degrees. The other quadrants can be derived by working back through the table for 90-180, and inverting the values for 180-360.



Please tell me how to find values for a 256-points Sine Table as you suggested above? I am trying to find the maths to generate this table but cannot find it so far.
#18
Jim Nickerson
User 452
  • Total Posts : 5578
  • Reward points : 0
  • Joined: 2003/11/07 12:35:10
  • Location: San Diego, CA
  • Status: offline
Re: DC/DC Boost Converter 2018/10/18 09:33:12 (permalink)
0
Don Lancaster explains sine wave calculation https://www.tinaja.com/magsn01.shtml
#19
du00000001
Just Some Member
  • Total Posts : 2238
  • Reward points : 0
  • Joined: 2016/05/03 13:52:42
  • Location: Germany
  • Status: offline
Re: DC/DC Boost Converter 2018/10/18 11:50:47 (permalink)
+3 (3)
Please tell me how to find values for a 256-points Sine Table ...

 
It's unbelievable! You should stick to 3.3 V logic as you might electrocute yourself during this project.

PEBKAC / EBKAC / POBCAK / PICNIC (eventually see en.wikipedia.org)
#20
Page: 123 > Showing page 1 of 3
Jump to:
© 2018 APG vNext Commercial Version 4.5