Controlling AC powered devices

Thread Starter

Khisraw

Joined Nov 14, 2020
56
Hi All,
As an electronics engineer, I have always worked with low voltage circuits and found it very difficult to control high voltage, AC powered devices. In my last project, I needed a power board that I can just plug my MCU or DSP and control an instrument that needs high power.
I wanted an AC/DC board to produce me 5V DC to power up my MCU and also have an AC output that can be controlled with a single from my MCU. After lots of searches, I learned that there is nothing in the market.

Have you had the same problem? I noticed a lot of competent programmers, and electronics engineers struggle with such issues. Please share your experience.

I have created a board that can generate 5V DC and has an AC output pin that can be controlled with the signal from your digital device.

What do you guys think? I need some feedback.
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
Can you give some definitive answers?
Depends what you call H.V. I control 240vac devices with a microprocessor using a opto isolator and a Mosfet to operate such as a relay etc. Or just operate the Mosfet direct.
Or to control a SCR or PWM bridge for H.V. DC motor control
Max.
 

Thread Starter

Khisraw

Joined Nov 14, 2020
56
Can you give some definitive answers?
Depends what you call H.V. I control 240vac devices with a microprocessor using a opto isolator and a Mosfet to operate such as a relay etc. Or just operate the Mosfet direct.
Or to control a SCR or PWM bridge for H.V. DC motor control
Max.
Control 240vac or 110 Vac devices with a microprocessor using an AC/DC controller.
 

Ian0

Joined Aug 7, 2020
9,667
The triac is your friend!
Most MCUs will drive a sensitive gate triac directly, so all you need is a -5V supply (with a few hundred mV of ripple) from a resistive or capacitve dropper circuit from mains neutral. Connect VDD of your microcontroller to mains live, and Vss to the -5V supply and you can control dozens of mains outputs and lots of power! If you have any switch inputs, make sure that they are will insulated!
There's a big discussion here https://forum.allaboutcircuits.com/threads/switching-a-24vac-solenoid-triac-relay-or-ssr.173864/ on the relative merits of Triacs and Solid-state relays, opto-triacs, isolated and non-isolated supplies. Most of the same arguments apply to mains as they do to 24V.
 

MisterBill2

Joined Jan 23, 2018
18,167
The suggestion from Max for a solid state relay is far safer than a triac, because the good quality SSR devices have well separated terminals that are well marked. And they have very clear ratings.
In addition, I have never seen a COMPETENT electrical engineer struggle with such things. I have seen a few technicians who were NOT competent struggle quite a bit, though. Electrical engineering and writing control code are two very different activities needing different skill sets. The fact that some few engineers can write good code is a testimony to their being able to learn, it is a different skill set indeed.
 

Ian0

Joined Aug 7, 2020
9,667
I'd say that it pays to learn how to use a triac safely. Cheapest 1A SSR in RS is £3, cheapest 1A triac is 7p ; cheapest 10A SSR is £17, cheapest 10A triac is 12p. A lot of difference between 12p and £17.
I don't see the SSR ratings being any clearer than the triac ratings. Both have peak voltage ratings, single cycle current ratings and RMS current ratings, and minimum trigger currents.
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
In addition, I have never seen a COMPETENT electrical engineer struggle with such things
The suggestion from Max for a solid state relay is far safer than a triac, because the good quality SSR devices have well separated terminals that are well marked. And they have very clear ratings.
Although if one is an electrical engineer as stated, I would think it should be well within their capability. ;)
Especially if a lowly Industrial Electronics Technician can successfully achieve it. :D
There are many app notes such as Fairchild AN-3006 and AN-3009 for more info.
Max.
 

Reloadron

Joined Jan 15, 2015
7,501
All I used was primarily 3 phase SSR blocks and occasionally relays/contactors slaved to smaller driver relays. That included a wide range of pumps, heater elements, motors up to about 50 HP and other devices running 480 VAC 3 Phase Delta power. All of it from 5 volt logic out of assorted uC devices, PLC and other process control systems. I don't see where there is a problem?

Ron
 

MisterBill2

Joined Jan 23, 2018
18,167
My point in that comment was because it is easy to have good isolation between the controls computer and the AC mains with that 17$ SSR, while it takes a bit of circuit building to isolate the 5 volt logic from the mains when using a triac. Certainly it can be done, but not as simply. AND the Crydom-style SSR devices mount very nicely on a steel panel that serves as a heat sink, and they do not need any mica washers.
 

BobTPH

Joined Jun 5, 2013
8,804
I have used opto isolated triacs that can handle 1A purchased for about a buck. They have the advantage of switching at zero cross. Came in an 8-pin DIP.

Bob
 

Ian0

Joined Aug 7, 2020
9,667
I have used opto isolated triacs that can handle 1A purchased for about a buck.
The standard opto triac (MOC3063 etc) can drive a small load by itself. It can also be used to switch larger triacs.
Crydom-style SSR devices mount very nicely on a steel panel that serves as a heat sink, and they do not need any mica washers.
All the Thomson triacs with an A in the part number (BTA08-600TW, for instance) are isolated and don't need micas. The ones starting BTB are non-isolated.
With both the isolated triacs AND the solid state relays, when mounting on an earthed chassis, one has to watch the creepage and clrearance distance between the leads and the chassis, especially on a 230V supply.

By the way, I've nothing against SSRs, I'm just putting the case for the humble triac. One thing in its favour was replacement costs. When controlling filament lamps (remember them?) triacs often failed when the filament of a large lamp failed. If the customer was faced with a replacement cost of 50p versus £17 for a solid-state relay, he would have been none too happy.
 

MisterBill2

Joined Jan 23, 2018
18,167
This discussion was started by an individual who claimed to be struggling with the way to achieve control. That tells me that neither an 8-pin dip device nor a triac component would be easy for them to utilize. So the suggestions that are useful depend quite a bit on the abilities and skill sets of the ones doing the posting. Not all of the participants are skiled tech folks or experienced engineers. We often see introductions claiming to be "newbies" who have questions. The SSR devices that I suggested are the very simplest ones to use and physically some of the most rugged ones available.
 
Last edited:

s14rs4

Joined Sep 15, 2016
75
This discussion was started by an individual who claimed to be struggling with the way to achieve control. That tells me that neither an 8-pin dip device nor a triac component would be easy for them to utilize. So the suggestions that are useful depend quite a bit on the abilities and skill sets of the ones doing the posting. Not all of the participants are skiled tech folks or experienced engineers. We often see introductions claiming to be "newbies" who have questions. The SSR devices that I suggested are the very simplest ones to use and physically some of the most rugged ones available.
I totally agree with you. The initial hardware might be expensive, but the implementation is simple. If you fuse the output and use an adequate heatsink it is virtually unbreakable. I have one here that is driving an 8A load at 240VAC. The input is directly from an Arduino Nano output, you don't even need a series resistor, it is current limited from 3V- 32V.

Also something no one has mentioned is SSR's are available with zero-crossing switching which eliminates switching noise and interference.
 

Ian0

Joined Aug 7, 2020
9,667
This discussion was started by someone who claimed to be an electronics engineer. That implies to me someone who would have no difficulty whatsoever in soldering a TO220 package to a circuit board and fastening it to a heatsink.

SSRs are very useful components, but there seems to be some very specious claims about how fantastic they really are. All SCR devices - thyristors and triacs - are very reliable - a SSR is no more reliable than the triac that is inside it!
No-one has mentioned zero-crossing opto-triacs either.
Zero-crossing switching may eliminate interference, but it's no good for switching a big transformer, it just guarantees that it will saturate at the end of the first cycle and take a huge current spike, whereas random phase switching would actually make that less likely.
 

MisterBill2

Joined Jan 23, 2018
18,167
@Ian0, note that the brands of SSRs that I have mentioned are the better ones, not the cheap junk knock-off poor copies that are also available. amazingly enough, good quality products are much less likely to fail than those made with poor quality parts.
AND, if a system is going to be fused, it is better to fuse it on the power inlet side, so that all of the pieces are protected.
 

s14rs4

Joined Sep 15, 2016
75
@Ian0, note that the brands of SSRs that I have mentioned are the better ones, not the cheap junk knock-off poor copies that are also available. amazingly enough, good quality products are much less likely to fail than those made with poor quality parts.
AND, if a system is going to be fused, it is better to fuse it on the power inlet side, so that all of the pieces are protected.
You are correct, when I said to fuse the output I was referring to the power side of the SSR not the output of the SSR.
 

MaxHeadRoom

Joined Jul 18, 2013
28,617
If low power is all that is required, you can pick up the Opto22 for cheap. they come in a few configurations, AC,DC Input, Output 5v or 24v dc control.etc.
Boards come in 4,8,16, 32 , they have built in LED and fusing.
They can produce a PLC style system when using a Picmicro etc.
I have a couple of buckets of them going surplus. ;)
Max.

1611073628674.png
 

MisterBill2

Joined Jan 23, 2018
18,167
Thanks for the reminder max, that name had slipped my mind. Certainly it is a good quality brand, and I think that there are interchangable brands available as well. Thus availability should not be a problem.
 
Top