Test Ideas: Sort thyristors with a test circuit
Use this circuit to sort for latch current and hold current.
By Chee Hua How, TDK Malaysia -- Test & Measurement World, 2/1/2010 2:00:00 AM
![]()
|
Thyristors conduct current from their anode to cathode (IAK) when they receive a gate pulse of sufficient current and duration. They will continue to conduct if IAK reaches the latching current (IL) level even after the gate pulse ends. The device should remain on as long as IAK exceeds the device’s hold current (IH). If IAK falls below a minimum IH level, the thyristor should turn off.
Thyristor data sheets will specify typical values for IL and IH, but you may need to sort thyristors based on the difference between IL and IH—not their individual values—to ensure that the thyristor will properly control a circuit. If you ignore the variation of these two parameters, the thyristor may cause a product failure should it fail to properly turn on or off. Thus, you may need to set limits that guarantee proper circuit operation.
![]() Figure 1. You can use this circuit to sort thyristors based on IL(MAX) and IH(MIN). |
The circuit in Figure 1 lets you set values for IL(MAX) and IH(MIN) that will ensure proper product operation. IL(MAX) and IH(MIN) are two extreme properties that never coexist in a single device. You can then use LEDs to indicate the thyristor’s pass or fail condition. Table 1 describes the four possible test results.
The settings in Figure 1 are for screening a TSN10A80 thyristor to fulfill requirements of IL ≤ 20 mA and IH ≥ 6 mA. The circuit uses a 20-mA current source formed by U1 and Q3. This current can then travel through the LED1 branch, through the LED2 branch, or through the series RC circuit (R9 and C3). A dual monostable device, U3, sets two clock pulses (CLK1, CLK2) that trigger either the DUT or Q5, provided the pulses are of sufficient length to guarantee that the respective devices turn fully on and off.
|
Start a test by pressing switch S1, which forces Q2 off and thus forces Q1 off. That momentarily removes VCC (20 V) from the circuit (Figure 2). Following a short delay tDLY1, the first firing signal, CLK1, goes to the DUT (device under test) to turn it on. The DUT will conduct the entire 20 mA from U1, which will illuminate LED1. Q4 and Q5 will not conduct because CLK2 is low.
After period t1, U3A will remove the gate pulse CLK1 from the DUT. A blanking time of tDLY2 will follow. The RC network R11 and C4 set the blanking time, which is the period in which the DUT should latch in its conducting state. If the DUT fails to latch, it won’t conduct current and LED1 will extinguish (condition 1 of Table 1).
If the DUT latches, IAK will be 20 mA and U2B will turn on, sending a trigger pulse to its triac side, U2A. After tDLY2 elapses, U3B will initiate CLK2 and Q5 will conduct for period t2. Q5 should then latch in its “on” condition. When Q5 activates, the U4-Q4-U2A network will divert 14 mA from the DUT branch. If the remaining current (6 mA in this example) of DUT current is below the DUT’s holding level, LED1 will extinguish and LED2 will remain on (condition 3 of Table 1). If not, the DUT has failed to turn off and both LEDS will remain on (condition 4 of Table 1).
The values shown in Figure 1 are based on measurements for the TSN10A80 thyristor. To use this circuit with another thyristor, you will need to adjust the duration of tDLY1, Δt1, tDLY2, and Δt2 by changing the values of R18•C6, R16•C5, R11•C4, and R17•C7, respectively, to change time constants. Proper timing of t1, tDLY2, and t2 will ensure that the current flowing into the DUT reaches steady state before the firing signals (CLK1 and CLK2) time out.
|
Proper timing also prevents excessive overshoot or undershoot current. The CLK1 and CLK2 pulse durations in Figures 3 and 4 show 105 µs and 120 µs, respectively (100 µs/div). The tall waveform in Figure 3 shows a failed DUT, while Figure 4 shows a passed device. In both cases, the DUT successfully latched (tall trace). In Figure 3, IDUT dropped to 6 mA after CLK2 activated, diverting 14 mA through Q4 and Q5, but the DUT remained on. In Figure 4, IDUT dropped to 0 mA, indicating that the DUT had shut off. Thus, the device tested in Figure 3 had a value of IH(MIN)≤6 mA, but in Figure 4, IH(MIN) > 6 mA.
TABLE 1. Test result state diagram
| Condition | LED1 | LED2 | Result |
| 1 | Off | Off | DUT fails to latch |
| 2 | On | Off | Invalid; tester problem |
| 3 | Off | On | Pass |
| 4 | On | On | DUT fails to turn off |
-
Here is my reply to Mr Bauerle comments
(1) Figure 1 is incorrect - pins 5/11 of the 4538 are the active-low inputs and should have the dots on them, not pins 4/12.
(1a) Nice spot by Mr. Bauerle. Yes he is right there are errors on the symbolic representation of U3A input (pin 11 should be dot [-TR] and pin 12 [+TR] should have no dot) and U3B input (pin 5 should be dot and pin 4 should have no dot) of Fig.1.
(2) The timing figure is thus incorrect too – U3A triggers on the low-going edge of the 20-V.
(2a) As per datasheet pin 11 is (HIGH-to-LOW, edge-triggered input) therefore U3A is still triggers on low-going edge of 20V.
(3) The feedback from pins 10 and 6 also doesn’t serve to lock-out retriggers.
(3a) Pin 10 is Q2 output whereas pin 12 is +TR input, by connecting them together as Fig.1 the lock-out (non-retriggered) mechanism will be ensured.
(4) D5 should be reversed and fed from pin 9, and pins 4/5 should be swapped to take advantage of the Schmitt trigger input of U3B. Likewise, pins 11 and 12 should
be swapped.
(4a) In general the main confusion comes from my symbolic representation of +TR and –TR of U3A & U3B (as explained in 1a). Once the symbol is redrawn it should be ok. The pin numbering of U3A and U3B still remain ok (matched with IC functionality), provided that the circuit is constructed according to the IC pin numbering as per fig. 1 statement (2), (3) and (4) are no longer valid. Sorry for confusion that caused due to my drawing and Thanks to Mr. Bauerle for addressing it out.
Chee Hua, How - 2010-24-3 06:31:38 EDT -
Here is my reply to Mr Bauerle comments
(1) Figure 1 is incorrect - pins 5/11 of the 4538 are the active-low inputs and should have the dots on them, not pins 4/12.
(1a) Nice spot by Mr. Bauerle. Yes he is right there are errors on the symbolic representation of U3A input (pin 11 should be dot [-TR] and pin 12 [+TR] should have no dot) and U3B input (pin 5 should be dot and pin 4 should have no dot) of Fig.1.
(2) The timing figure is thus incorrect too – U3A triggers on the low-going edge of the 20-V.
(2a) As per datasheet pin 11 is (HIGH-to-LOW, edge-triggered input) therefore U3A is still triggers on low-going edge of 20V.
(3) The feedback from pins 10 and 6 also doesn’t serve to lock-out retriggers.
(3a) Pin 10 is Q2 output whereas pin 12 is +TR input, by connecting them together as Fig.1 the lock-out (non-retriggered) mechanism will be ensured.
(4) D5 should be reversed and fed from pin 9, and pins 4/5 should be swapped to take advantage of the Schmitt trigger input of U3B. Likewise, pins 11 and 12 should
be swapped.
(4a) In general the main confusion comes from my symbolic representation of +TR and –TR of U3A & U3B (as explained in 1a). Once the symbol is redrawn it should be ok. The pin numbering of U3A and U3B still remain ok (matched with IC functionality), provided that the circuit is constructed according to the IC pin numbering as per fig. 1 statement (2), (3) and (4) are no longer valid. Sorry for confusion that caused due to my drawing and Thanks to Mr. Bauerle for addressing it out.
Chee H. How - 2010-24-2 08:02:30 EST -
Figure 1 is incorrect - pins 5/11 of the 4538 are the active-low inputs and should have the dots on them, not pins 4/12. The timing figure is thus incorrect too - U3A triggers on the low-going edge of the 20-V. The feedback from pins 10 and 6 also doesn't serve to lock-out retriggers.
D5 should be reversed and fed from pin 9, and pins 4/5 should be swapped to take advantage of the Schmitt trigger input of U3B. Likewise, pins 11 and 12 should
be swapped.
Ron Bauerle - 2010-22-2 16:58:53 EST -
Why doesn't the PDF button work?
LD - 2010-12-2 07:09:45 EST
No related content found.
- 0 rated items found.
Datasheets.com Electronic Parts & Inventory Search
185 million searchable parts
- Part Number
- Description
- Inventory
- Products
- Manufacturers

























