Talk:TRIAC

Gates

The details begin with "gate", and it appears as if there is only one gate at first. But then, in the descriptions and examples that follow, there suddenly appears to be two different gates - namely "gate 1" and "gate 2". So, the newcomers would be confused. How many gates are there for a triac? One? Two? Furthermore, the illustrations appear to be erroneous, in that the quadrant 3 diagram says "gate 2 negative", but the diagram currently has the "gate" on MT1 (Main Terminal 1). A similar issue for the quadrant 4 diagram, which says "gate 1 positive", but the "gate" is connected to MT2 according to the quadrant 4 diagram.

Those diagrams also shows four (not three) nodes. So why show four nodes when triacs are generally sold with three pins? — Preceding unsigned comment added by KorgBoy (talk • contribs) 09:33, 12 December 2016 (UTC)

I think I can help here. All of these devices have 4 or more layers and a nominal Cathode / Anode arrangement. In an SCR the "gate" is the base of one of the virtual transistors nearest the Cathode. In an SCS, both "gates", or the bases of both virtual transistors, is brought out at both the Anode and Cathode. In a PUT, the "gate" is the base of the virtual transistor nearest the Anode. In this case, there are two more PN junctions, creating two more virtual transistors as well as a fifth virtual transistor at the "gate". In the introductory schematics, the diagrams are the same, but inverted. The "gate" is the same terminal whether named "gate 1" or "gate 2". I'd recommend putting a caption on the diagrams in that section to clarify the inversion and removing the "1" and "2" from the "gate" designation. 50.247.247.81 (talk) 14:54, 19 September 2017 (UTC)

The drawings make no sense any way you turn them. It is written gate+/gate- and an `equivalent` circuit with transistors are shown, but the equivalent circuit has more pins than a triac does... It is also not mentioned in respect to what the gate is + or -. Voltages always require two points. Have a look at this pdf from ST and you will quickly see how confusing the image given in the wiki is. https://www.st.com/content/ccc/resource/technical/document/application_note/ce/e4/e4/c5/d3/bf/4c/c7/CD00003868.pdf/files/CD00003868.pdf/jcr:content/translations/en.CD00003868.pdf

After reading the wiki it is not even remotely possible to deduce that a triac could be controlled as shown in the st pdf. — Preceding unsigned comment added by 2A02:810B:4C0:66F6:A64C:C8FF:FE1C:9E42 (talk) 02:32, 3 March 2019 (UTC)

I agree with the objections above. Figure 1 doesn't make sense: neither is it correlated to the triggering mechanisms presented in the "Operation" section, which can be verified for instance in source [2], nor is it consistent with other literature, such as the AN https://www.st.com/content/ccc/resource/technical/document/application_note/ce/e4/e4/c5/d3/bf/4c/c7/CD00003868.pdf/files/CD00003868.pdf/jcr:content/translations/en.CD00003868.pdf.

Therefore, I am going to revert this figure to the following one, which I originally posted 8 years ago and was found in this article as late as this article's revision https://en.wikipedia.org/w/index.php?title=TRIAC&direction=prev&oldid=503427601.

• 

It looks amazing to me that a wrong picture has remained on this article for such a long time. Vitaltrust (talk) 08:54, 2 November 2019 (UTC)

As the author of the diagram you removed, and as the electronic engineer who has extensively used these devices and physicist who researched these at the quantum level, I can state that the diagram is an obvious extension of the "equivalence curcuit" presented in quandrant 1.

• 

• 

So please be far less subjective in your criticism. It would help if in the spirit of wikpedia you improved it instead of replacing it with a meaningless image table of not real use to anyone who is attempting to understand these devices. Grahamatwp (talk) 20:43, 2 February 2021 (UTC)

For what it's worth, I find the removed illustration perfectly sensible, useful, and indeed obvious, and I see no reason to remove it. It is quite clear from Figure 2 that the two gates (one being N-type and one P-type) are in fact shorted together.

(...) the quadrant 3 diagram says "gate 2 negative", but the diagram currently has the "gate" on MT1 (Main Terminal 1). A similar issue for the quadrant 4 diagram, which says "gate 1 positive", but the "gate" is connected to MT2 according to the quadrant 4 diagram.

I don't see any place in the diagram where gate (whether "gate 1" on the left or "gate 2" on the right) is connected to a main terminal.

After reading the wiki it is not even remotely possible to deduce that a triac could be controlled as shown in the st pdf.

I assume this was coming from someone who failed to realize that there is in fact only a singe gate (and thus the TRIAC is a three-terminal device), and the two gates are connected together. Once you manage to understand this, such "deduction" should be quite straightforward. 90.183.96.145 (talk) 01:02, 8 January 2023 (UTC)

(Not same user as below.) As someone using this wiki page to assist in actually using a TRIAC for the first time, i managed to realise that the GATE 'terminal' (on the outside) somehow connects internally to 2 separate gates of two different "virtual" transistors. However, i have trouble imagining a proper equivalent circuit, since when you connect the gates of the PNP and NPN "virtual" transistors together (and assuming the emitter of the PNP is more positive than the emitter of the NPN), then the TRIAC would always be in the on-state, as base current could then flow from PNP's base, through the common terminal, and into NPN's base, immediately switching on both transistors even if the GATE terminal were left unconnected. Is there a more detailed equivalent schematic that addresses this issue?83.132.147.130 (talk) 11:47, 20 August 2024 (UTC)

Note from WP:NOTHOWTO that this isn't a user's manual, or otherwise telling how to use them. You should find the actual data sheet for a specific model. Otherwise, you have to be careful when considering virtual transistors. Much easier to consider the Silicon controlled rectifier first. Since they aren't actually two separate transistors, the operation happens much faster. That is important as much power is dissipated during the switch on time. Otherwise, turn-on is tricky. You might need to supply the right gate voltage while supplying power to be sure it doesn't come on. Then there is dV/dt, where a fast changing voltage, and internal capacitance, will turn it on. Gah4 (talk) 07:34, 22 August 2024 (UTC)

Thanks for your reply. In the meantime i was able to come up with an equivalent circuit that completely reproduces the TRIAC's switching behaviour in all 4 quadrants (ie, all triggering modes). I guess there's no point in posting it as it would be original research, though i actually feel it enlightens the matter considerably, especially for those like myself having little experience with thyristor-like components.83.132.230.13 (talk) 03:12, 25 August 2024 (UTC)

Once you manage to understand this, such "deduction" should be quite straightforward.

This is quite the silly remark when the context is a wikipedia article, a place where you'd expect to learn such information. One can't deduce said context from the image without learning about the aforementioned and by you mentioned three-terminal nature of the TRIAC. Unfortunately that requisite information is misrepresented or misimplied exactly by the misleading image in question.

TLDR: I find it ironic for you to remark "You don't understand Y because you don't understand X. So you can't complain" while the entire premise of this Wiki page was to learn both X and Y.

You can't do that, especially not for subjects that are considered non-intuitive or not common sense (as is the case for electronics). Remember, what's common sense to you is common sense to you. Others don't have the context of experience and knowledge; they came here to learn said information, rather than be told they just misinterpreted it all due to the user not being smart enough. We are supposed to be people teaching the basics to those who don't know them yet, after all? 2A02:1811:B11A:9850:6563:2D90:BDC4:9E09 (talk) 15:04, 30 April 2024 (UTC)

TRIAC patent

Hi to all, I found this link: http://patentimages.storage.googleapis.com/pdfs/US3275909.pdf US Patent 3 275 909 by Frank W. GUTZWILLER Filled on Dec./19/ 1963 Patented on Sept.27 1966 92.152.102.56 (talk) 15:14, 15 April 2017 (UTC)link found by Jacky THIELLIN -France

References

Other names

Symistor--Юе Артеміс (talk) 13:00, 16 March 2018 (UTC)

Triacs aren't used for induction motors

Article indicates triacs can be used to control speed of induction motor. Though one might be able design a circuit that does this to prove a point, one cannot (generally speaking) control the speed of an induction motor by adjusting the voltage it gets. Torque, maybe. But to control speed you need to control frequency, not voltage, and lowering voltage under load usually leads to overheating. Reference to induction motor should be replaced with brushed ac or "universal" motor. The wiki article for universal motors affirms that an advantage of such motors is easy speed control via triac. — Preceding unsigned comment added by 73.162.24.160 (talk) 07:17, 7 March 2019 (UTC)

Triacs don't control voltage, but instead timing of turning it on. (Off at the next zero crossing.) As far as I know, it is more usual to use SCRs for big motor control as the current is higher. There is a favorite circuit for controlling a universal motor with an SCR, which uses the back-EMF to control the SCR triggering. TRIACs are an old favorite for light dimmers, though. Gah4 (talk) 23:18, 2 February 2021 (UTC)

Continuing this, yes, induction motors run based on input frequency. Given a DC power supply, and (usually) SCRs, you can run an induction motor, switching them on at the right time. For example, Seattle's Link light rail is powered by 1500 VDC power. It is converted to three-phase at the desired frequency, I suspect with SCRs and not TRIACs, but the idea is the same. The high switch speed, microseconds, of SCRs and TRIACs generates RFI in the AM radio band. You can easily hear this on a car radio, driving near the Link trains. The design is slightly complicated by the need for each to switch off at the right time. Gah4 (talk) 07:22, 22 August 2024 (UTC)