1. Home
  2. Electronics Design
  3. Triggering large SCR modules and hockey pucks with continuous DC gate current

Triggering large SCR modules and hockey pucks with continuous DC gate current

This relates to the 12V DC-DC converter previously posted. I am designing a new SCR trigger board similar to previous versions that had proven unreliable with pulsed gate drive and with DC gate drive only during the forward conduction mode. The load is highly inductive, and can be heavily loaded for high current, or mostly inductive and low current into the primary of a large transformer with open secondary.

The problem is due to the phase shift between current and voltage. The initial firing phase angle is set to about 70 degrees, which usually produces the least amount of DC offset and an initial half-cycle most closely matching the subsequent half-cycles. The old design alternated drive based on the reference AC voltage zero crossings, so the gate current was actually maintained only for a portion of the current waveform, and when the current was below the hold current, the device would go out of conduction and cause much distortion and high current surges.

So we found that an initial gate signal followed by continuous DC worked well to achieve good waveform under all conditions. I designed the original boards around 1980, and the gate current has been supplied by a constant current source set at about 300-350 mA. This has proven to be reliable for large hockey puck SCRs rated 1600-1800 volts and 500-2000 amps, as well as smaller modules rated at 600-1600 volts and 90-160 amps. The SKKT162/16E modules specify a maximum required gate turn on current of 150 mA and pulse current up to about 10 amps. Minimum current is about 350 mA at -40C and 150 mA at 25C. Here is a document on gate drive, mostly for larger devices (DC1596SW):

formatting link

Here is a datasheet of what is probably a similar device, 1800V, 1700A:

formatting link

That spec sheet shows gate current of 300 mA nominal, and 150-300 mA, although there are points on the curves from 200 mA to 4 A, with gate power of 20W maximum. It seems that a higher trigger voltage requires less current. My latest SCR board uses about 11 VDC at which point it appears that 100 mA is sufficient. The previous SCR boards had an open circuit voltage of about 10 VDC and dropped to 6 VDC under load.

The new board produces about 200 mA continuous with a somewhat higher initial surge. That may help to trigger the initial turn-on at a specified phase angle more accurately. The board draws about 550 mA when initiated, or about 6.6 watts. I am considering changing the transformers to have half as many secondary turns to get a 6 VDC open circuit voltage with 300-400 mA at about the same input power.

Any comments, suggestions, and experience are welcomed.

Thanks,

Paul

Reply to
P E Schoen
Loading thread data ...

Some SCRs, especially big ones, need a fast, hard initial trigger. A weak trigger can fire part of the chip, which can burn out before the PNPN trigger action spreads out over the whole chip.

So one useful waveform is a big initial spike, many amps, followed by some DC to hold it on as required.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com
Reply to
John Larkin

This is an AC switch, with two SCRs connected in anti-parallel. So the initial pulse could have a waveform with a fast rise and a higher current peak, but thereafter, the DC will be maintained on both devices. With an inductive load, the initial current will start low and then rise with a sinusoidal waveform delayed by up to 90 degrees. When subsequent half-cycles occur, the gates will already have been on for at least a quarter cycle or 4 mSec. So by that time I think the gate structure will be fully turned on.

Perhaps the minimum gate current can be determined by applying a gradually increasing current until it fires, and do so for several samples. Then perhaps twice the current of the highest sample would be safe.

Thanks,

Paul

Reply to
P E Schoen

I got the SCR gate characteristics with the data sheet for the large SCRs we are using. They are rated 1800V and 960A average:

formatting link

I understand that the gate current and voltage will relate to temperature, and a higher voltage will be needed when it is cold, but I am unsure about the higher voltages shown in the graphs. The lower limit appears to be the typical diode junction voltage of 0.3 to 0.7 volts, but I am surprised to see voltage as high as 3 volts at 300 mA and 4 volts at 500 mA.

Figure 13 shows a rather large recommended trigger area, bound by the 20 watt maximum gate power, but I don't understand the left side which shows

200 mA from 10 to 12 volts, and 300 mA at 8 volts, and finally about 1 volt at 900 mA. The previous boards had at least 12 VDC open circuit (actually almost 20 VDC), but then current limited to about 300-350 mA.

That may be more for a pulsed gate trigger. The device turn-on delay is given as 3 uSec, with a gate source of 30 volts and 10 ohms.

Full specs:

formatting link

Paul

Reply to
P E Schoen

Actual specs:

formatting link

Paul

Reply to
P E Schoen

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.