Hello All,

I would very much like to hear from people who can assist with the following:

1. What is considered to be the 'typical' Total Harmonic Distorion (THD) produced by a soft starter?

2. What infomation is required to enable accurate calculation of THD for soft starter applications?

3. Has anyone experienced problems arising from Harmonic Distortion created by soft starters?

Looking forward to your responses.

Regards,
GGOSS

GGOSS,

We recently did some testing for a customer re: harmonics. We used one of our 6 SCR Phase Angle power controllers since we could hold any particular firing angle long enough to measure THD. What we found was the worst THD was found right at 50% firing angle, at about 20% current distortion, under 5% voltage. Most distortion was found in the even orders, 2nd, 4th, 6th, with some noticeable levels in the 5th and 11th. THD steadilly improved as the firing angle was advanced, and was barely detectable at 95%.

If one were to consider that most applications start at ±50% initial voltage and average ramp time is under 30 seconds it is my opinion that the transient nature of the event does not warrant mitigation. I have not had anyone challenge that statement as of yet. Here in the US we are frequently required to meet IEEE 519 specifications for THD, and that spec makes no allowance for time-limited events. I still have not had a problem when I provide this description to a reasonable engineer.

This of course goes out the window when a feature such as the energy saver or voltage rebalancing is used where the SCRs firing is retarded continuously. Unfortunately since we discontinued that feature in our products I was unable to get this added to the test proceedure, so I have no hard numbers on that.

Many thanks jraef, this is great information. However I would have thought that the even order harmonics created by a 6 SCR system would be very low, primarily because this gives symetrical control of both positive and negative portions of the waveform.

Are you able to comment on why the even order harmonics were so high?

Regards,
GGOSS

GGOSS,
Since we have other engineers who are more specialized in matters like this, I was not directly involved in the testing other than as a liason to the user. I did get the impression that they (the other engineering group) were a little surprised by this as well, but since "why" was not at issue for the test they did not do any further analysis.

I'll ask now.:P

Hello jraef

I am likewise very surprised to hear that you found such a high proportion of even harmonics. This is what I would expect from an SCR Diode starter, but not an SCR SCR starter. I wonder if this was a digital based design, or an older analog design. From my experience, the even harmonics are very small relative to the odd harmonics. (based on measurements made)

Best regards,