Design of Permanent Magnet DC Motor Drive

[senzafine]

Hey there i would like help to design a permanent magnet dc motor Drive.

The drive has the following components:
48V Dc motor
Mosfet H-bridge inverter
Single-Phase Rectifier
Mains Transformer
Dc Link Capacitor
Gate Drive Circuit
Control Electronics (inc. microcontroller)

The motor: Brushed Permanen magnet field on stator.
Armature Voltage: 48V
Rated Speed: 3000rev/min
Rated torque: 0.146 Nm
Rated Curren: 1A
ktψ: 0.146 Nm/A
Ra: 2 Ohm
La: 10mH


The difficulties I have: is 1) I don't know how to choose the correct MOSFETS for the H-bridge . Also it is said that a 48V Dc link would probably be sensible. 2) I don't know how to choose a suitable Dc link capacitor. Because it says that i have to choose it based on the dc link voltage with a safety margin. 3) I am struggling to calculate the maximum electrical power input to the motor when it delivers full rated mechanical input and i have to calculate this in order to choose a suitable bridge rectifier


I really need your help. Please if someone can help me is for a design project i have to do for the university. Thanks

 

[senzafine]

I know that generally for dc motors Pmec = Pel + Plosses but for this specific purpose Pmec = T *ω but is it the rated value that i am going to use or the max value ? also about the losses Plosses = I*I*R in that case is it the rated current or again i have to calculate the maximum curren? Please if anyone can help it would be really helpfull

 

[KeepItSimpleStupid]

You HAVE to be able to supply 48V into a 2 ohm load because that's what the system sees when it starts. t=0+. This is also the locked rotor current I=48/2 Now, i don't know how long you have to be able to sustain it, but that can be determined by any protection mechanism you might employ. The simplest would be a little larger than the expected blow time of a fuse.

Any FETS selected for the bridge should be based on SOA or Safe Operating Area. Pulse response will be there too.

Again, the surge ratings of the bridge and the power supply also has to deliver the voltage for longer than this time period.

 

[senzafine]

Dear KeepItSimpleStupid,

Thanks for your answer. To be honest i am thinking to divide the circuit into three stages. 1) Input Stage , 2) Inverter and 3) Control . I know that i have to include in my calculations for the input stage the losses i will have due to the dc link capacitor the diodes . But i have to select the ratio of the ac /dc transformer i am going to use. after calculations i have chosen 1/7 ratio of the transformer. What is your option about that ?

 

[senzafine]

sorry i have wrote wrong the value of the ratio is actualy 6.75:1 so i said that if i add a 10% allowance for losses then i will have to use a 7,4:1 transformer ratio. Do you think its correct the assumption i am doing please if anyone can help me

 

[KeepItSimpleStupid]

Take a look here: https://www.electro-tech-online.com/custompdfs/2012/10/5c007.pdf It's a brief design guide. DC linear power supply design was a nice chapter in my EE classes.

Note it says to take into account Vd and other losses. One spec is what AC power really is supposed to be in the US. I "Think" it's +-10% from 120 VAC, but not sure. Some of the "universal" power supplies actually specify voltage less than 100 VAC because of Japan and some higher than 277 VAC. I just looked at a computer brick and it says 100-240 VAC input which I would read as "nominal voltages".

Here https://www.digikey.com/product-sea...dc-converters/590377?k=universal power supply is one that specifies 90 to 264. 90 = 100-(~10%) and 264 = 240+24 or (~10%). So, that's where the specs come from. The Utility, not random.

Do, I know your mains voltage?

 

[senzafine]

Yes my supply is a single phase and because i am studying in the uk i used 230v in my above calculations sorry i didnt mention it but is my first time i am doing that project and i am really confused where to start and how to structure my steps . Thank you very much keepitsimplestupid. Can i ask something else in order to select a correct bridge rectifier i have to make sure that my diodes will withstand the voltage and current of my dc link with a safety margin again isnt it? Thanks again for your help

 

[KeepItSimpleStupid]

i picked a bridge which I would probably call "marginal" for your project. The 50V part. Here is a datasheet: https://www.electro-tech-online.com/custompdfs/2012/10/dfm.pdf

Now, I would probably use a 100 PRV diode, possibly even 200 PRV. You can check the surge rating by using all of the stuff you know. The largest contributor (the most effect) will be the capacitor's ESR. This is where knowing that you likely can neglect the winding resistance, but include the diode drops just to check the surge current.
Checking the surge current though may be more than the instructor requires.

Watch temperature, too.

You can refer to me as KISS if you'd like.

 

[senzafine]

Thats brilliant kiss thank you very much i will continue with the calculations of the mosfets where i am going to use and post them here in order to give me your opinion about them again thank you very much

 

[senzafine]

Hey guys i want to ask something. I have estimated the maximum electrical power input to the motor by using the formulae Pin = Pmech + Plosses which in my case is 0.146*3000*2π/60+1*1*2 = 47.86 Watts isn't this value small ? thanks

 

[senzafine]

Hello again, thanks for all the support provided till now . Can I ask something else about the Dc link capacitor. it says on the project requirements that the choice of voltage is fairly straight forward based on dc link voltage with a safety margin. and also it says I need to find out what safety margins are considered appropriate. I have done a research on the internet and I could find anything which is quite useful. can anyone help me about that bit ? and again thank you very much for everything

 

[KeepItSimpleStupid]

I would rate the capacitor and 2x the motor voltage. This isn't unusual.

I kinda have real world example na power amplifier that I built. The power supply is basically 50 V unregulated and I used two 100 uf 50 V electrolytic caps on the board itself. The supplies were filtered with computer grade 9600 uf caps on each of the 50 V supplies. This particular design used a CVT transformer and one day the 100 uf Caps blew catastrophically.

After that incident, the supply was changed to a 4 x 50 V DC with 9600 uF caps on each rail and the caps were upgraded to 100 V. I had another issue because LED's as part of an optocoupler would burn out that were basically across the caps. I had to place ZNR's across each cap.

The amp has effectively ~50,000 uF of capacitance on a 50 V supply when it powers up. I had to create a slow turn-on circuit to deal with that too.

 

[senzafine]

thanks for your quick reply is there any documentation i can find in order to verify the capacitor i am going to choose or i just have to choose it based on some calculations ? because if its based on the calculations i guess the current ripple must be calculated and then a capacitor according to that will be chosen. if i am correct

 

[KeepItSimpleStupid]

Ripple current is the right way.

A "rule of thumb" is 1000 uF/amp.

 

[senzafine]

Ok thank you very much

 

[KeepItSimpleStupid]

I found this for you: https://www.diyaudio.com/forums/pow...sizing-linear-unregulated-power-supplies.html

So, as you can see, your on the right track.

That article mentions efficiency, so I also found this: https://www.engineeringtoolbox.com/electrical-motor-efficiency-d_655.html

 

[senzafine]

Thank you very much it is a nice calculation program

 

[senzafine]

Hello guys again i would like to ask a question in order to calculate the diode of the bridge rectifier i know that my secondary voltage after the transformer will be 48v . how do i calculate the conduction or the switching loses. i know switching are very small because the frequency of supply is 50hz but again does it worth it calculating them in order to find out how much drop i will have in my circuit ? . I know that in a bridge rectifier every half period 2 diodes are conducting therefore i am having 1.4v drop every half period so if i say that for a whole period it will be 3v voltage drop is correct or wrong ??

 

[alec_t]

so if i say that for a whole period it will be 3v voltage drop is correct or wrong ??

Wrong. Two diodes of a bridge conduct in one half-cycle; the other two conduct in the other half-cycle. So only 1.4V drop in both half-cycles.

 

[senzafine]

Thank you.
Do you believe it worth's to analyse the conduction and switching losses for both the rectifier diodes and mosfet ? or just take the voltage drop for rectifier and analyse the conduction for mosfet ? thanks again for your help