1

Firstly I am sorry for this bad drawing but I just got it from the internet.

Secondly I'll give a brief idea about the project it is a static switch would be used to drive general loads (inductive and resistive mostly) unfortunately I have no idea about the loads types and values but mostly will draw 10 A maximum 220 Vac such as heating elements or motors.

Third I will explain the changes that I made:

I changed R1 from 220 Ω to 10 Ω because when I first connect the circuit to live power (without connecting the optotriac to 5 Vdc (Moc 3061 ZVC I am using)) the SCR gets triggered for a fraction of second and gives a pulse of 311 Vac peak (220 Vac RMS)when I changed it to 10 Ω surprisingly it became much more stable and done that for R3 too.

R2 got changed from 500 Ω to 100 Ω because when I changed R1 and R3 to 10 Ω R2 got burned so I lowered it to 100 Ω.

I added a snubber circuit too (R4 and C1)

Now the questions are:

  • Is this circuit considered to be sufficient for general using in the Market?

  • what is the use of these two diodes?

  • what is the general snubber circuit is used in Market (for general purpose I need it because I know it should be calculated for specific loads)?

  • Is it possible to just use a TVS diode instead of a snubber?

Note: I had tested it and it worked fine (I tested it for short duration).

If there is any information that you need please consider to ask in the comments.

Edit: Link of the circuit: Original picture

enter image description here

MiNiMe
  • 1,609
  • 5
  • 19
Hazardous Voltage
  • 185
  • 9
  • 4
    Sure, but why not use a triac? Also you reversed the SCR's anode and cathode labelling. – 15 Volts Oct 28 '23 at 04:51
  • 1
    Did you add the snubber before or after changing the resistors? The resistors values are not permitted: check the maximum ratings of the MOC. – Tim Williams Oct 28 '23 at 05:53
  • I must misunderstand something about this circuit? The current through R2 is about 2 A RMS (220 V / 110 Ohm) causing a power dissipation of 400 W in R2? – Oskar Skog Oct 28 '23 at 07:18
  • 2
    @OskarSkog Presumably an SCR fires, shorting out said voltage drop; evidently this didn't happen, after changing R1/R3 but before changing R2(!). (You've also assumed no load resistance, which isn't a great assumption, but, operation into a short-circuit load might indeed be a good consideration; one would hope for adequate fusing, for example.) – Tim Williams Oct 28 '23 at 07:58
  • 1
    IF you are looking for a cct to use in volume production for general purpose use then this circuit is not suitable. This is a "lashup" circuit probably produced by someone who does not have a good idea of the subject and using two SCR's for no apparent reason when one TRIAC would work better. The circuit appears on many websites. The original designer MAY have had some reason to do what they have done BUT, if so, it is not evident what it was. || If you want a general circuit for volume production look art many 1 x TRIAC based circuits and look for one whose design is described. – Russell McMahon Oct 28 '23 at 08:56
  • Oncve you have one or a few candidate circuits ask about it/them. Ideally find a circuit used in quality equipment made by a reputable brand of equipment. There are various of rthos on the internet. – Russell McMahon Oct 28 '23 at 08:58
  • 1
    @15Volts the Triacs won't be quite suitable due to excessive heat generated which would require big heatsink on the other hand with the SCR I can easily use it with a smaller heat sinks. – Hazardous Voltage Oct 28 '23 at 16:06
  • @TimWilliams I added it after changing the resistors and it worked – Hazardous Voltage Oct 28 '23 at 16:06
  • @OskarSkog as I had measured on the oscilloscope the peak of voltage across R2 is approximately 7 Vac RMS so it won't burn. – Hazardous Voltage Oct 28 '23 at 16:08
  • @TimWilliams I agree with you 100% for that I asked for the snubber and TVS and in this circuit I just didn't draw the fuse but there would be a 15A fuse and need a TVS or snubber too. – Hazardous Voltage Oct 28 '23 at 16:11
  • @RussellMcMahon the problem with one Triac is that it gets quite hot and it would need bigger heatsink too – Hazardous Voltage Oct 28 '23 at 16:17
  • @RussellMcMahon can you recommend a brand or if you have a link for such circuit would be quite helpful? – Hazardous Voltage Oct 28 '23 at 16:19
  • 2
    TRIACs drop little more voltage than SCRs do. Single units in common packages (e.g. TO-218, TO-247) are perfectly usable up to 30A. You need as much total heatsinking with a pair of SCRs, but twice the power components, and probably insulators (heatsink hardware in general). You can also consider a ready-made AC type SSR, which is the same circuit internally, and has a wide heatsink base to bolt down. – Tim Williams Oct 28 '23 at 16:46
  • 3
    FYI to all - the original source of the schematic, with a little more background, seems to be here (and then adapted & copied to other places, including the Edaboard post mentioned by the OP). – SamGibson Oct 28 '23 at 16:51
  • @TimWilliams so you are saying that if I used a SMD triac package would it be able to handle 10 A of power without a heatsink? – Hazardous Voltage Oct 28 '23 at 20:28
  • I'm not sure where you (intentionally?) misunderstand, when I list specific, larger, THT packages and you get "SMD" from that. – Tim Williams Oct 28 '23 at 21:19
  • @TimWilliams My apologies mate I had misread it as SMD packs like the TO-263. – Hazardous Voltage Oct 28 '23 at 21:41
  • ADDING TO MY ABOVE COMMENT: The original circuit and design is here - the aim was to reduce heat per package by using an SCR for each alternate half mains cycle. The total heat is about the same. – Russell McMahon Oct 29 '23 at 00:30
  • @RussellMcMahon yes you are definitely correct because at 220 Vac with a 1.73 A load the triac reached just near the 100 C (I was using Bta12) the SCRs (Bt152) reached about 50 C or less but the heatsinks I would be using will be much smaller than the triac. – Hazardous Voltage Oct 29 '23 at 01:10
  • 1
    @RussellMcMahon Responding to "..using two SCR's for no apparent reason when one TRIAC would work better". || When I used SCRs and TRIACs many decades ago now, IIRC, there was a significant difference in the dv/dt they could withstand across A-K without causing false turn-on, with SCRs having much higher values for this parameter than TRIACs. Note, I have not checked the datasheets in recent times to confirm this. Perhaps this may have been part of the original motivation for this circuit. Cheers. – Fabio Barone Oct 29 '23 at 20:26
  • 1
    @FabioBarone Thanks. We now know that his aim is lower heat dissipation & smaller heatsinking AND that that was the aim of the original design. I ued Digikey selector guide to choose an SCR and a TRIAC in 100 quantity 20A, 400A, TO220, lowest cost. SCR had perhaps a 10-20% thermal edge over TRIAC in that case. Two SCRs cost about 33% more than the TRIAC. || TRIAC : SCR – Russell McMahon Oct 30 '23 at 06:35
  • I used Digikey selector guide to choose an SCR and a TRIAC in 100 quantity 20A, 400V, TO220, lowest cost. | SCR had perhaps a 10-20% thermal edge over TRIAC. Two SCRs cost about 33% more than the TRIAC. || Data sheets: TRIAC : SCR ||After mounting and heatsink issues etc the difference is not (in my opinion) worthwhile. Seleted device (in either case) may be better. – Russell McMahon Oct 30 '23 at 06:38
  • @RussellMcMahon I agree with you about the pricing especially the PCB would be bigger – Hazardous Voltage Oct 30 '23 at 14:20
  • 1
    @RussellMcMahon Thanks, referring to those two devices you suggested, interesting to see that both seem to have identical dynamic characteristics: Critical Rate of Rise of Off−State Voltage of 100V/us (min), and maximum permitted rate of rise of on-state current = 50A/us (although the specific test conditions vary). Interestingly, maximum permitted rate of change of commutating voltage was given for the TRIAC and not the SCR. Cheers. – Fabio Barone Oct 30 '23 at 22:37
  • 1
    @FabioBarone Thanks. Your examination was more detailed than mine. As noted, I used (voltage & current) and then lowest price and then looked at Von and similar. The idea was to get a rough feel for differences. It's interesting that the two parts match so well in the areas that you note. || If actually specifying a part I would of course have gone into other aspects in more detail. "Lowest price" is not always a good metric :-) – Russell McMahon Oct 31 '23 at 23:03
  • @RussellMcMahon Agreed - "pay half, pay twice" :-)
    || I was just curious as to whether my previous experience with TRIACs & SCRs regarding their differences in dynamic behaviour was still applicable today - these two examples were surprisingly very similar. Cheers.     – Fabio Barone Nov 01 '23 at 03:12
  • @FabioBarone I have a question mate if you may, now I don't understand why did you suggest that we increase the resistor values of 10 and 100 ohms, I mean are not they safe or would they have low lifespan? and about the Opto-coupler the LED side is already turned off when I connect the circuit to AC power. – Hazardous Voltage Nov 01 '23 at 11:49
  • 1
    My commentsabove are unchanged. The TOTAL dissipation with 2 x SCR OR 1 x TRIAC is about the same. The total heatsinking is about the same. Both are rated at ABOUT 1.5 K/W so the TRIAC definitely gets hotter due to continuous conduction. If your situation is so tightly constrained that your heatsinking reduction gained is a major issue then you are probably making a marginal product. || Q1 IMPORTANT - WHY is heatsink minimisation so crucial for you? – Russell McMahon Nov 01 '23 at 21:07
  • @HazardousVoltage I suggest looking at the datasheet application notes, for example:|| https://www.onsemi.com/pdf/datasheet/moc3163m-d.pdf Page 6 shows the "inverse back-to-back" circuit in your OP, they suggest making R1=R3 (your diagram) = 330ohm, and R2=360ohm, and note that this is for a 115V AC supply, a 220VAC supply would need higher values again to limit the current to the same values. – Fabio Barone Nov 01 '23 at 21:21
  • @HazardousVoltage I have updated my answer to respond to your question, cheers. – Fabio Barone Nov 01 '23 at 21:42

1 Answers1

6

I will attempt to answer your questions here. Firstly, some background context collated from the "comments" section (thanks to Sam Gibson, Tim Williams, & Russell McMahon):-

Source of the circuit:

  1. As provided by the OP:
    https://www.edaboard.com/threads/back-to-back-scr-driver-circuit.401988/

  2. Original source: Refer Fig 4 of this link, image copied below:
    https://www.bristolwatch.com/ele3/1d.htm

enter image description here

Responding to your questions:

Is this circuit considered to be sufficient for general using in the Market?

No. There are many things missing, just for starters: over-load and short-circuit protection (eg: fuses), over-voltage protection, EMC filters, safety eg: electrical isolation.

what is the use of these two diodes?

Referring to D1 & D2:
(a) They prevent reverse voltage being applied to the G-K terminals of each SCR.
(b) They ensure that the gate trigger current is determined by R2 & the load, and is not affected by R1 & R3.

what is the general snubber circuit is used in Market (for general purpose I need it because I know it should be calculated for specific loads)?

As you indicate, the exact snubber depends on the intended application, the type of load, and the details of the SCRs used. For example, if the mode of operation is to trigger the SCRs only when the applied 220VAC voltage is zero (zero-crossing) then then snubber required will have a lower rating than for random-trigger point-on-wave triggering (where the trigger can occur at any point on the AC cycle). Futhermore, if you intend to use this circuit for dimming lamps (where the trigger may occur at any point on the 50Hz waveform, and is repeated for each AC cycle), then this means the snubber will be working at 100Hz or 120Hz (once in the positive half, and once in the negative half) and this will cause the highest power loss within the snubber (and the SCR). It will also generate the most EMI - which needs to be carefully filtered, the snubber is a part of that EMI filter.

Loads can be resitive, inductive, or capacitive; a purely resistive load will probably have the lowest-rated snubber, whereas a capacitive load will require a different type of snubber compared to an inductive load.

Snubbers are used to:
(a) protect the SCR from over-voltage, & over-current (particularly at turn-on into capacitive loads);
(b) prevent false turn-on (usually caused by high dv/dt across A-K), and
(c) limit power dissipation within the SCR for non-resistive loads.

Please note that different types of SCRs have different values and limits for all these parameters, and these parameters vary with temperature, so the snubber must be designed for the specific SCR being used (and the manufacturing tolerances to be expected), and the operating conditions - ambient temperature, and internal temperature rise.

Is it possible to just use a TVS diode instead of a snubber?

Possibly, but it depends - see the answer above.

Snubber design:
These links may be a good starting point:

Snubber design for switching an inductive mains load with a relay?

How to design an RC snubber for a solenoid relay driving an inductive load??

Designing an RC Snubber for TRIAC motor control?

Update 30-Oct-2023:
Responding to this part of the OP:

Third I will explain the changes that I made:

I changed R1 from 220 Ω to 10 Ω because when I first connect the circuit to live power (without connecting the optotriac to 5 Vdc (Moc 3061 ZVC I am using)) the SCR gets triggered for a fraction of second and gives a pulse of 311 Vac peak (220 Vac RMS) when I changed it to 10 Ω surprisingly it became much more stable and done that for R3 too.

R2 got changed from 500 Ω to 100 Ω because when I changed R1 and R3 to 10 Ω R2 got burned so I lowered it to 100 Ω.

I added a snubber circuit too (R4 and C1)

What you experienced here is known as "the first pulse effect": how the circuit behaves when power is first applied. This behaviour becomes more complex as the number of power sources increases; in this case there are two: (1) the AC supply, and (2) the voltage at the LED side of the opto-coupler. The RC snubber you added has a large effect on this behaviour, and you may find that this snubber has corrected the false triggering that occurred on the first pulse with the resistors R1, R2, & R3 set to their original values.

I would suggest changing resistors to higher values; 10Ω & 100Ω seem too low to me, and may result in excessive gate currents. Check the SCR data sheet. Don't forget also that the optocoupler output stage has to carry this current.

I would also suggest that you take into consideration the "first pulse effect" for the opto-coupler, and consider using resistors to ensure the LED is kept off during the start-up phase; in particular, you must ensure that currents caused by capacitive coupling between nodes with high dv/dt cannot cause unintended turn-on.

Update 02-Nov-2023:
Responding to part of the OP's comments under the OP:

@FabioBarone I have a question mate if you may, now I don't understand why did you suggest that we increase the resistor values of 10 and 100 ohms, I mean are not they safe or would they have low lifespan? and about the Opto-coupler the LED side is already turned off when I connect the circuit to AC power. – Hazardous Voltage 9 hours ago

My response: Because the current flowing in the opto-TRIAC may be too high. May I suggest looking at the datasheet and application notes. For example, assuming the opto triac you are using is the MOC3061TVM, and its datasheet is here:
https://www.onsemi.com/pdf/datasheet/moc3163m-d.pdf

Page 6 shows a typical application for firing two back-to-back SCRs (image below):
enter image description here

This suggests making R1=R3 (using the component IDs of your diagram, not those of the diagram above) = 330ohm (or not installed at all, these are "optional"), and R2=360 ohm, and note that this is for a 115V AC supply, a 220VAC supply would need higher values again to keep the current to the same values.

Suggest consider the two cases:

  1. When the opto LED is OFF (no current). Look at the voltages present on all the components of your circuit.
  2. When the opto LED is ON (conducting forward current). Look at the currents that are flowing in all the components of your circuit.

The other cases to consider are:
3. The transition between case 1 (OFF state) and case 2 (ON state).
4. The transition between case 2 (ON state, when the main SCRs are conducting load current) and case 1 (OFF state).

For cases 3 & 4 you need to consider in small time steps how the circuit behaves as it makes these transitions. This is when the snubber, and how it interacts with the load, becomes very important. A simulation program, such as LTspice, will be very useful for that.

Fabio Barone
  • 2,101
  • 1
  • 17
  • unfortunately mate, when I used the snubber with a higher value resistors( 330 ohm instead of 10 ohm) the capacitor of the snubber used to charge and discharge in the circuit causing more pulses so for that I needed to decrees the resistors and it worked and I don't know if it would be safe or not – Hazardous Voltage Oct 30 '23 at 14:16
  • Fun fact: you are right when I lowered the resistor value and drew excessive load from the triac wheres the temperature raised more than 140 degrees, then I turned off the circuit and turned it on again the optotriac blew out with a quite big spark however I don't know if the resistors were the main reason for such behavior. – Hazardous Voltage Oct 30 '23 at 14:18
  • If the opto-TRIAC you are using is the zero-voltage-switching type (the output turns on only at the zero-crossings of the mains voltage, then beware: driving an inductive load with this can cause problems. Suggest drawing up the expected load current waveforms for a purely inductive load to see what I mean. Cheers. – Fabio Barone Nov 01 '23 at 21:46