Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

3phase filter

Status
Not open for further replies.

furyfax

New Member
Hi, I've been given a task to design a low pass filter. (Passive filter, no OP-AMPS).
I only want 0-60Hz to pass.
Look at my fantastic mspaint drawing, thats the system. I'm going to have filter at the grey lines.

So far I've chosen R = 10KOhm and C= 0.27uF.
Then Fc will be about 59Hz..

But I'm not that familiar with designing filter, so I just want to know if the info I've given you guys seems to be OK?

There are 3 phases, can I have one filter at each phase or won't this work ?
 

Attachments

  • filter.JPG
    filter.JPG
    17.5 KB · Views: 208
Presumably (from the diagram) this is for a motor?, in which case you can't use an RC filter, you need an LC one. But why would you want a filter to a motor?.

You should also be aware that a simple filter like this doesn't 'stop' out of band signals, it merely reduces them, and not by very much just out of band.
 
From what i understand the motor doesn't cope with the noise which comes to it. Therefore a filter is needed.
I've been told to make some kind of filter (low-pass), only 0-60Hz is OK.
 
right so you are more after a EMC filter right? (as opose to signal feedback's for control?)

yes machines don't like HF high current...current (increase eddie loss and also increase bearing wear as well as something todo with current's going doe the bearings...) hence when power inverters started really comming onto the scene the life of machines actually came down

So the question is do you know the power density and spectrum you need to reject? this AINT a quick LC will-do the trick thing
 
Styx said:
right so you are more after a EMC filter right? (as opose to signal feedback's for control?)

yes machines don't like HF high current...current (increase eddie loss and also increase bearing wear as well as something todo with current's going doe the bearings...) hence when power inverters started really comming onto the scene the life of machines actually came down

So the question is do you know the power density and spectrum you need to reject? this AINT a quick LC will-do the trick thing

There's 690V AC and U_peak= 1200V DC, and I believe max ampers is 0.1A. My english isn't that good, sorry, don't know if I understood u correctly.
 
Anyone got any good links to LC filter design ?
I'm struggling to satisfy the specs I've been given.

I'm more a fan of programming than filter design.:mad:
 
furyfax said:
Anyone got any good links to LC filter design ?
I'm struggling to satisfy the specs I've been given.

What specs have you been given?.

Bear in mind the filter has to pass the full motor current, so it's likely to be large and expensive. I would imagine, if this is a normal requirement?, that such filters would be freely available?.
 
The motor is controlled by frequency transformer. 0-60Hz is the area. 60Hz is full speed.
From what I've been told there's a lot of noise at 1500Hz.
So the main task is to get rid of this noise, but idealy a filter which just lets 0-60Hz through would be best.

When I've tried to make a LPF, I get sick values on the capacitors..
This isn't my area, so I'm pretty fresh when it comes to filter design.
-------------------------------------------------------------------
EDIT:
I've misunderstood the task somewhat. The 3 phases go to the motor but also a transformer. I thought we were to have the filter infront of the motor, but it's infront of the transformer.
So then I won't need to take the motor in consideration.
 
Last edited:
Nigel Goodwin said:
I think if you try looking at those figures you will find they don't make any sense at all?.

Yes... ;)

I realize I've misunderstood key elements in the task..
The filter should be in front of a transformer were I have to remove the 1500Hz noise and I don't need to take the motor into consideration..
 
FuryFax, I understand you want a 60hz low-pass filter that can handle 0.1 amp and 690 volts AC with peek at 1200 volts. I think your real goal is to remove 1500hz noise.

1. A 60hz filter is BIG (pounds or Kg). A 600hz filter will remove the 1500 noise and it will be 1/10 the size.

2. Your voltage is AC+DC. Does the voltage ever go negative? This effects what types of capacitors we can use. I have Panasonic 1600 VDC 0.056uf capacitors. Paralleling two gives 0.1uf.

3. Coil: https://www.electro-tech-online.com/custompdfs/2007/08/5c0032.pdf 1.5H 200mA coil

4. 1.5H, 0.1uf = 400hz low pass filter (200mA 1600 volt)

5. Build three filters. One for each leg. If that works try removing one of the filters.
 
What's driving the motor? Direct AC mains? Or some sort of variable speed drive (that switches)? If it's the mains you probably can get away without a filter. If it's a variable speed drive that uses PWM (ie. switches somehow), you will need one.

Although, I noticed you said frequency transformer. Did you mean to say variable frequency drive? Or frequency converter? But the way you say how the filter will be placed in front of the "frequency transformer" doesn't make sense if this is actually the case since it should go between the motor and drive to filter out the switching harmonics. Or do you actually have a transformer that somehow is able to convert between frequency (do those even exist?!!)

Sometimes you don't need the capacitor, only the inductor, due to those "sick capacitor values".
 
Last edited:
It is clear we don’t see the whole picture. None of us have seen a motor and motor driver like this. The motor changes speed with frequency (strange) maybe it is a stepper motor. Maybe there is a 1500hz PWM involved. Probably none of the above is true.
 
ronsimpson said:
The motor changes speed with frequency (strange) maybe it is a stepper motor.

Or it is a sync machine...
be it BLDC or BLAC or even rotating diodes

Or it could even be... an Induction machine working with V/f control.

Plenty of machines change their speed w.r.t. electrical frequency so it aint strange, it is fundemental
 
Is this a modified sine wave inverter?

I take it the problem you're having is that the odd harmonics are causing excessive heating in the motor as they're reverse clocked?

Won't a normal harmonic filter do?

I'm sure you can buy these off the self.
 
I don't think he knows what he's got, or what he wants?, these were the specs he gave:

The motor has an impedance of 0.0018 Ohm.
U= 690V AC
U_peak= 1200V DC
The current is maximum 0.1A to motor

Which doesn't make much sense with ohms law?.
 
Hero999 said:
Is this a modified sine wave inverter?

I take it the problem you're having is that the odd harmonics are causing excessive heating in the motor as they're reverse clocked?

Won't a normal harmonic filter do?

I'm sure you can buy these off the self.

Thats what I am thinking and yer an off the shelf part is usually best unless something special/specific being done
 
Nigel Goodwin said:
I don't think he knows what he's got, or what he wants?, these were the specs he gave:



Which doesn't make much sense with ohms law?.

Cause Mr Ohm's law isn't as straight forward when it comes to electrical machines, esp when you start factoring in: inductance, saliancy, backEMF...

he *may* have a winding resistance of 0.0018 Ohm. and he *may* only need 0.1A to flow. The only thing questionable abt the numbers given is that amp level is far to low to be useful, a 690Vac machine will have a few hp behind it.

What is more likely is those figures ARE correct, its just he has not explained what he has been asked todo correctly.

the main machine is probably going to be drawing a few 10's maybe 100's of amps and the source of it is noisy. Some form of conducted immisions testing was probably done and it was shown that there was 0.1A of HF conducted current that is unwanted. THIS is what needs to be filtered out

So he *may* be needing an EMC to filter out some switching noise? and others from freq > 60Hz.

Still lacking alot of info admitidly
 
Styx said:
Still lacking alot of info admitidly

If I ever need a screenplay writing from a single line of random words?, I know where to come from! - your ability to create a reasonable sounding post from meaningless values is astounding! :D :D :D
 
Ha!

its just I deal with motor-drivers and EMC filters are part of it.
bits of what he has posted all point to an EMC filter to remove HF current from the main current seen by the machine.

I am hoping it joggs him into going "yer thats what I want and this is the other info"

but seems he aint returning
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top