Soft Starters and RFI

[theDOG]

I am currently working on an application that has three soft starters (of the same make) at different pump stations. When the pumps are starting, there is severe radio interference on a particular AM radio station in the area. This diminshes when the pump is at speed, but is not totally eliminated (the soft starter is not bypassed). The pump stations are quite a distance from the supply transformer, and residential properties are also connected to the same supply.

 

Question 1.

Can anyone please explain how the radio interference is caused?

 

Question 2.

How can it be rectified? Can we simply add some impedence to the input ie. line choke. Or would a common mode core be more suitable?

 

Question 3.

Would all soft starters behave the same?

 

Cheers.

[marke]

Hello TheDOG

 

This is a common problem in remote rural areas!

 

>Question 1.

>Can anyone please explain how the radio interference is caused?

 

The radio interference is caused by the action of the SCRs switching. Any cahnge in voltage and/or current will emit an electromagnetic disturbance. The severity and frequency range of the disturbance is dependant on the rate of change of voltage and current, and the magnitude of the change. The annoyance level is related to the magnitude and the repetition frequency of the disturbance. In the case of a soft starter, the repetition rate is 100Hz which is much less than a modern drive at up to 15KHz.

The rate of change of voltage and current is relatively slow compared to modern inverters. The modern IGBT's used in drives switch in 100 - 200 nS as opposed to somewhere in the order of 5 - 20 uS for SCRs. The result is that the disturbance levels are much lower and only cover a restricted frequency range compared with modern inverters.

The major energy switching occurs during start and soft stop, during RUN the SCRs should be controlled at vertually zero volts accross them and so the emissions should be negligable, certainly very low compared to the emissions during start and stop.

Soft starters employing energy saving features produce RFI during RUN because they are reducing the voltage and therefore switching with higher rates of change of voltage and current.

EMC requirements generally accept the burst during start, but do not tollerate the continuous emissions during RUN from the energy saving algorithms. Devices are only approved provided the energy saving is disabled.

 

>Question 2.

>How can it be rectified? Can we simply add some impedence to the input ie. line choke.

> Or would a common mode core be more suitable?

The problem is not always easy to cure because there are a number of parameters that affect the generation of the RFI.

The motor and cabling to the motor have a relatively high capacitance to ground, and so when the SCRs switch, you are providing charging current to the capcitance. This has a high di/dt and therefore high emmission. Additionally, the SCRs are protected by snubber networks that comprise capacitance and the charging current into these snubber networks is also a source of high di/dt.

The third problem, is that unlike inverters, there is no isloation of the switching elements from the supply, so any conducted RFI is passed directly on to the supply. Having said this, the levels are much lower than the equivilent inverters and filters are rarely necessary. Tests that I have been involved in for CE complience has not required any filters.

The best way to reduce the RFI, is to reduce the rate of rise of current di/dt and really, this is something that you need to reffer to the equipment manufacturers for comment. Standard series filters will help, but I have not found any filters that will "fix" the problem.

A bypass contactor will solve the running problem.

Note: there may also be a problem from some units due to radiation by the onboard microprocessor of switchmode power supply. These would tend to create problems while the unit is online but the SCRs not operating.

 

>Question 3.

>Would all soft starters behave the same?There will be some variation between soft starters due to variation in the snubber design and construction, also the presence of some degree of RFI filtering in some units. The major difference will be the behaviour while the unit is running.

 

Best regards and good luck,

;d;

[GGOSS]

Hey theDOG,

 

Good to see you active again even if its because you want us to solve your application problems for you

 

I recall a similar experience at a pumping station within a residential zone many years ago. Believe it or not, the solution was to purchase a new AM radio for the little old lady that complained about the RFI during pump starting and stopping ie the new radio solved the problem.

 

Not the most technologially advanced solution but it worked for us. Give it a go!

 

Regards,

GGOSS

[marke]

GGOSS

Good point.

There are two components in EMC, the emissions and the susceptability. It is very possible that there are two problems!!

;c;

[GGOSS]

Hey theDOG,

 

Any joy with the new radio?

 

Regards,

GGOSS

[theDOG]

Hi Everyone;

 

Sorry for not keeping you up to date.

The problem has been fixed by simply installing a 3%z line reactor. We tried to use a ferrite core per phase, with little success (it helped, just not enough!).

 

Cheers!

[marke]

Hello theDog

 

Pleased to hear that you sorted the problem.

Did you fit this on the input side or the output side?

Did you try both configurations?

 

Best regards,

[theDOG]

It was fitted on the input. We didn't try it on the output,as space was limited.

[GGOSS]

Many thanks for coming back to us with advice on remedial action taken and results achieved. This is great information for everyone involved in the wondeful world of soft starters and beyond;d;