Scr Firing Methods Difference
Posted 30 August 2006 - 05:31 PM
I have a question about SCR Firing methods in a Medium Voltage Soft Starters. Some manufacturers use a DC continues firing signals (about 170 degrees length), another -Economy Pack Pulses (about 5 us pulses and 5 us pause etc...) Explane me pls what difference between these 2 methods? What is the best method applying to high voltage softstarters?
Posted 30 August 2006 - 06:58 PM
There is a posting about this in the SCR section of this forum.
2) Picket Fence Firing
3) Trigger Pulse firing.
In theory, with an SCR, you can apply a single pulse to turn the SCR ON and it will remain ON until the current through the SCR falls to zero.
In practice, there are a number of constraints in using the Trigger Pulse firing.
The duration of the trigger pulse must be longer than the time taken for the current through the SCR to increase to a current higher than the latching curent of the SCR. If the current falls below the holding current of the SCR, the SCR will self commutate. This can cause premature commutation if there are disturbances in the current flow due to supply or load disturbances.
The use of Hard Fire techniques overcomes thses potential issues. Hard fire techniques requires that the gate current is held ON for the duration of the conduction angle of the SCR. This ensures that the SCR can not be prematurely commutated. This requires gate energy and presents a higher load to the trigger circuit and extra heat in the SCR. It is a little difficult to generate a trigger pulse of 10ms (6.66ms for 60Hz) that has a fast risetime and requires more than just a pulse transformer.
Picket fence firing comprises a series of pulses applied to the gate for the duration of the expected current flow. These short pulses are easy to generate with a pulse transformer and ensure that the SCR can not commutate OFF for a period of time greater than the pulse repetition rate. It is a compromise between the Trigger pulse technique and the Hard Fire technique.
When you then consider the MV applications of soft starters, you have an additional complication of several SCRs in series in each phase. If any one SCR in a phase should comutate OFF early, the whole chain is interrupted, plus there is a very high probability that a very high voltage will be presented across that one SCR until the rest of the SCRs in that phase switch OFF. This increases the risk of SCR failure due to overvoltage and care must be taken to ensure that the voltages are balanced across the SCRs during their OFF state.
By far the most reliable method of controlling SCRs in series, is the continuous fire or Hard fire technique. Any form of strobed gated fire technique is a compromise which on single SCRs is not a major issue, is a potential problem with series connected SCRs.
Picket fence firing and single pules (short pulse) firing is used as a cost saver only. This should not be necessary in MV as the cost saved is small relative to the total cost of the starter.
Skype Contact = markempson | phone +64 274 363 067
LMPForum | Power Factor | L M Photonics Ltd | Empson family | Advanced Motor Control Ltd | LMP Software | Pressure Transducers | Smart Relay | GSM Control | Mark Empson Website | Soft Starters
Posted 01 September 2006 - 05:03 AM
Propagation of conduction.
If you were able to observe the silicon junction begin its transformation from insulator to conductor, you would see that it happens somewhat radially outward from the point where the gate is attached. Without sufficient energy in the gate pulse, the propagation wave takes too long, which makes the current flow concentrate on a smaller area of the wafer, overheating it. The Picket Fence method, while fine for low voltage systems, is not as effective in ensuring that high level of gate pulse energy necessary to prevent this. This can be exacerbated by the next issue mentioned below. Users of Picket Fence firing compensate for it by stringing together that rapid succession of pulses, but if the first one lit the wafer, the next one doesn't help much any more.
If there is any severe line noise present where the sub-cycle voltage actually dips below the zero cross, an SCR can actually commutate back off again. This is especially problematic at low firing angles. Picket fence firing attempts to "re-strike" the SCR gate, but unfortunately their pulse energy starts to diminish after the first one, so the likelihood of an incomplete fire goes up in that scenario. A "block pulse" firing method sends a very high energy long pulse to the gate, so IF the SCR commutates within the cycle, there is still a high amount of gate energy present to re-ignite it and finish the cycle with less stress on the SCR. By the way, some manufacturers use 210 degrees instead of 170 so that any ringing in the line can be conducted away. That's called a "DC back Porch carry over".
This is perhaps one of the best papers written on the subject that I have ever found and is widely regarded in the industry as the definitive guide.
Enerpro white paper on MV SCR firing topologies
0 user(s) are reading this topic
0 members, 0 guests, 0 anonymous users