Need advice on cheap SSRs

[Peabody]

I'm looking at starting a reflow project which may be a toaster oven or a hot plate.  But either way I'm probably looking at switching 10 amps, 110V mains.  Looking at various options for the solid state relay, I'm struck by the difference in price between cheap Ebay sources versus Arrow or Digikey.  The price difference is 5X or more for what should be the same thing - ratings wise.  The pictures on Ebay show "Fotek" SSRs, but in the detailed description they are all referred to as "unbranded generic".  So here's an example of a "40A" SSR complete with heat sink for US$7.80 delivered:

https://www.ebay.com/itm/152145657023

If Ebay is to be believed, they sell large numbers of these "Fotek" SSRs, so I have to assume they work most of the time.  But it's not clear what the difference would be between a good and bad relay other than the rating of the triac used for output.  And it's my understanding that when these things fail, they fail closed, so it's kinda important that they not fail. 

If I get one like the linked item, which is supposedly rated 4X the current I will actually draw, and has a heat sink, is that pretty likely to work out ok?   I mean, can I overcome the knockoff issue by selecting a 40A part and using a heat sink?

Anybody here have experience with these cheap SSRs?

Does anybody know of a source of cheap genuine SSRs that can be trusted to perform to their specs, not knockoffs of uncertain origin or quality, or do you just have to pay up to get a Crydom or whatever from Digikey?

[wraper]

Forget about them, they are steaming pile of counterfeit shit. Nor they use triac capable of supplying rated current, nor they are electrically safe. If you want something decent at this price, use triac + optocoupler.

[mvs]

If Ebay is to be believed, they sell large numbers of these "Fotek" SSRs, so I have to assume they work most of the time.  But it's not clear what the difference would be between a good and bad relay other than the rating of the triac used for output.  And it's my understanding that when these things fail, they fail closed, so it's kinda important that they not fail.

Fotek may produce safe products, but on ebay you will probably get a counterfeit product.
https://www.ul.com/newsroom/publicnotices/ul-warns-of-solid-state-relay-with-counterfeit-ul-recognition-mark-release-13pn-52/

[Zero999]

If Ebay is to be believed, they sell large numbers of these "Fotek" SSRs, so I have to assume they work most of the time.  But it's not clear what the difference would be between a good and bad relay other than the rating of the triac used for output.  And it's my understanding that when these things fail, they fail closed, so it's kinda important that they not fail. 

If that could be dangerous then your design is poor. Always have a thermal fuse to protect against fire, if there's a fault in the controller and the heating element is powered continuously.

As mentioned about, make your own SSR from a TRIAC and a zero crossing opto-coupler, if you want a low cost and safe SSR. Ensure the board creapage and clearances more than meet the minimum standards: a minimum 5mm gap between the non-hazardous and mains sides will be fine for mains voltages up to 250VAC.

[wraper]

Fotek may produce safe products, but on ebay you will probably get a counterfeit product.
https://www.ul.com/newsroom/publicnotices/ul-warns-of-solid-state-relay-with-counterfeit-ul-recognition-mark-release-13pn-52/

Yep genuine Fotek does exist but I wonder if anyone ever bought genuine on Ebay.

[Peabody]

Thanks for the responses.

If this project is ever completed, it will reside at a local makerspace.  With that, I would not feel comfortable trying to build an SSR from scratch.  So it may be that a $45 SSR from Digikey is the only option.  But you know, it doesn't really make sense that an SSR should need to cost twice as much as the entire toaster oven.  It seems there should be some source of good stuff at prices substantially less than Crydoms.

If Fotek is actually a legitimate brand, is there a place where you would be certain of getting their genuine stuff?  Are there any other low-cost (but not junk) SSR brands?

[Zero999]

Thanks for the responses.

If this project is ever completed, it will reside at a local makerspace.  With that, I would not feel comfortable trying to build an SSR from scratch.  So it may be that a $45 SSR from Digikey is the only option.  But you know, it doesn't really make sense that an SSR should need to cost twice as much as the entire toaster oven.  It seems there should be some source of good stuff at prices substantially less than Crydoms.

If Fotek is actually a legitimate brand, is there a place where you would be certain of getting their genuine stuff?  Are there any other low-cost (but not junk) SSR brands?

Does it have to be a solid state relay? If the switching frequency is very low and it's not subject to lots of vibration, use an ordinary relay. It will be cheaper and certainly no less reliable than an SSR, which has some shortcomings, compared to an ordinary relay: higher on losses and less electrically robust.

[Peabody]

Thanks for the responses.

If this project is ever completed, it will reside at a local makerspace.  With that, I would not feel comfortable trying to build an SSR from scratch.  So it may be that a $45 SSR from Digikey is the only option.  But you know, it doesn't really make sense that an SSR should need to cost twice as much as the entire toaster oven.  It seems there should be some source of good stuff at prices substantially less than Crydoms.

If Fotek is actually a legitimate brand, is there a place where you would be certain of getting their genuine stuff?  Are there any other low-cost (but not junk) SSR brands?

Does it have to be a solid state relay? If the switching frequency is very low and it's not subject to lots of vibration, use an ordinary relay. It will be cheaper and certainly no less reliable than an SSR, which has some shortcomings, compared to an ordinary relay: higher on losses and less electrically robust.

It will probably turn on and off once every 5 seconds for 4-5 minutes.  Think of it as a very slow PWM, with the duty cycle controlling the average power.  So probably less than 100 times per reflow.  But of course that's just one board, and someone may need to do several.  I don't really have any feel for how much it will be used, but every kit I see for this calls for an SSR.   I assume that's because switching 10 amps at other than zero crossing points would cause lots of noise, and wouldn't be so great for relay contacts.

[langwadt]

Thanks for the responses.

If this project is ever completed, it will reside at a local makerspace.  With that, I would not feel comfortable trying to build an SSR from scratch.  So it may be that a $45 SSR from Digikey is the only option.  But you know, it doesn't really make sense that an SSR should need to cost twice as much as the entire toaster oven.  It seems there should be some source of good stuff at prices substantially less than Crydoms.

If Fotek is actually a legitimate brand, is there a place where you would be certain of getting their genuine stuff?  Are there any other low-cost (but not junk) SSR brands?

this Panasonic  is 15A for ~$25 https:
https://www.digikey.com/product-detail/en/panasonic-electric-works/AQJ212V/255-5794-ND/4757810

[wraper]

If Fotek is actually a legitimate brand, is there a place where you would be certain of getting their genuine stuff?  Are there any other low-cost (but not junk) SSR brands?

There are, for example Cosmo KSD225AC8. What's available depends on where you buy. Don't expect below $20 for real 25A SSR.

[t1d]

I'm looking at starting a reflow project which may be a toaster oven or a hot plate.

I built a DIY reflow oven. I got a Crydom SSR relay, off Ebay. If you are willing to watch for a while, you can catch a decent deal, on new-old-stock (NOS.)

An advantage to SSR is that you can drive its control side directly from the controller pin. IIRC, my SSR control only required 7mA. I used a PIC18F4550. The project could likely have been done, with a much more basic chip.

My brother wrote the chip firmware and built a GUI, to control it, from a laptop, over USB. The project could likely have been done, with a much more basic chip, if you control the reflow program, from the chip, because the chip would not need USB capability.

The reflow code will need to drive the heating elements through a heating cycle. This cycle is defined by the solder and components that you use. To accomplish this, you will be cycling the SSR way more than every 4-5 seconds. You will also need PID correction. Look that up. And, the reflow cycle is only about 2-1/2 minutes.

I bought two identical quartz element toaster ovens, at the thrift store, and doubled the elements in one oven. Even so, our oven does not have a fast enough heating cycle. It also does not cool fast enough. This means that the heat lags behind the heat required at a given point in time, in the reflow cycle. And, it over-shoots the target heat, thereafter. Even so, our unit is usable and I use it all the time.

We have not published our project, but they are out there. See YouTube. You need to do lots of research, before starting.

And stay away from the Chinese Ebay SSRs. See Big Clive's tear-down/review, on YouTube.

[Zero999]

Thanks for the responses.

If this project is ever completed, it will reside at a local makerspace.  With that, I would not feel comfortable trying to build an SSR from scratch.  So it may be that a $45 SSR from Digikey is the only option.  But you know, it doesn't really make sense that an SSR should need to cost twice as much as the entire toaster oven.  It seems there should be some source of good stuff at prices substantially less than Crydoms.

If Fotek is actually a legitimate brand, is there a place where you would be certain of getting their genuine stuff?  Are there any other low-cost (but not junk) SSR brands?

Does it have to be a solid state relay? If the switching frequency is very low and it's not subject to lots of vibration, use an ordinary relay. It will be cheaper and certainly no less reliable than an SSR, which has some shortcomings, compared to an ordinary relay: higher on losses and less electrically robust.

It will probably turn on and off once every 5 seconds for 4-5 minutes.  Think of it as a very slow PWM, with the duty cycle controlling the average power.  So probably less than 100 times per reflow.  But of course that's just one board, and someone may need to do several.  I don't really have any feel for how much it will be used, but every kit I see for this calls for an SSR.   I assume that's because switching 10 amps at other than zero crossing points would cause lots of noise, and wouldn't be so great for relay contacts.

It's a resistive load, so I wouldn't expect it to be too noisy. There are plenty of other higher current, more noisy loads being switched in a typical house, office or factory.

I suspect the kits use solid state relays because the switching rate is quite high. I thought I'd have a quick look at mechanical relays, to see how long one would last in this application. The HF165FD/24-ZY1TF, which RS Components sell for £3.43. They don't have any stock at the moment, so you'll have to go else where, but this is just an example.

It's specified for 10⁵ operations with a load of 30A at 277VAC. If it switches 100 times per reflow, then you can expect it to do 1000 boards, but you're only switching 10A at 120V and the relay is specified for 10⁷ mechanical operations unloaded, so it will last longer than that.
http://www.hongfa.com/pro/pdf/HF165FD_en.pdf
https://uk.rs-online.com/web/p/non-latching-relays/1762835/

A solid state relay wouldn't have such a limit on the number of cyles.

[Peabody]

t1d, your post brings up several issues that I've been wondering about.

I have indeed spent a good bit of time watching the Youtube videos on all these toaster oven projects, and I still don't understand how they deal with the lag that must be involved with any temperature change. (You said your oven overshoots during reflow.)  I see everyone using feedback from the thermocouple as input for some kind of PID controller, but my understanding is that PIDs need to be *tuned*.  And with one exception, I don't see any provision for adapting the controller to the particular oven being used, or its level of added insulation, or added heating elements.

The exception I believe is the Controleo, which has a "Learn" mode.  It turns on the empty oven, then shuts it off at 100C, but then follows the temperature as it continues to rise, until it cools back down.  That information supposedly gives the controller the tuning information it needs.  But that controller is over $100, and I just don't have that kind of budget.

The other thing that strikes me is that anyone using the oven has to know how to place the thermocouple, because everything depends on that.

Given the variety of people who may be using the oven in my situation, with varying levels of understanding about the process, I came up with this idea for locking down the heating profile, and eliminating operator error:

I would establish a total ON/OFF cycle time of, say, 5 seconds.  For each such cycle the average power level would be determined by the duty cycle.  The 5 seconds would be divided by 100, giving duty cycle granularity of 50 ms.  The present setting would be shown on a 2-digit display, and would be set using a rotary encoder.

All the tuning would be done by me ahead of time.  Using a TM-902C thermocouple/display and a sample board, I would experimentally develop a sequence of duty cycle settings that produces the proper curve for leaded paste, such as:

10 cycles at 50%
14 cycles at 70%
etc.

As I develop the sequence, it would be loaded into the "controller", which would then simply execute that sequence when you push the button.  It would be the same sequence, and the same heating profile, every time.   The user can follow the temperature with the TM-902C if he wants to, but he doesn't have to.  And the thermocouple is not even connected to the controller.  There would be a buzzer indicating when to open the door.

Of course there's nothing magic about 5 seconds.  It could be anything that works.  Also, I'm not sure whether board size or parts composition would affect the reflow period, but as long as it doesn't overshoot, I hope a somewhat longer period should work for pretty much anything.  I'm also not sure about adjusting for a warmer initial temperature in case multiple boards are done.

I'll keep an eye out for an Ebay bargain on a Crydom or similar SSR.  And I may experiment a bit with a regular relay to see if that might work.

I'm also looking at the $8 Walmart Mainstays single-burner hotplate, with a steel circular saw blade on top.  In theory, it could be controlled by the same process.

[Brumby]

I'll keep an eye out for an Ebay bargain ....

That's the approach that could land you a counterfeit.  Just be wary.

[t1d]

I have indeed spent a good bit of time watching the Youtube videos on all these toaster oven projects, and I still don't understand how they deal with the lag that must be involved with any temperature change.

Well, that is the challenge.

I see everyone using feedback from the thermocouple as input for some kind of PID controller, but my understanding is that PIDs need to be *tuned*.  And with one exception, I don't see any provision for adapting the controller to the particular oven being used, or its level of added insulation, or added heating elements.

The PID calculation is what is intended to adapt the controller to the particulars of the oven. However, the PID controller can not overcome a lack of fast enough heat, or cooling.

Ovens with a convection air feature seem to do better, with heating and cooling speed. I am keeping an eye out for one at the thrift stores.

The exception I believe is the Controleo, which has a "Learn" mode.  It turns on the empty oven, then shuts it off at 100C, but then follows the temperature as it continues to rise, until it cools back down.  That information supposedly gives the controller the tuning information it needs.  But that controller is over $100, and I just don't have that kind of budget.

I researched these Diy-gone-commercial units and they all seemed to have design issues related to how much they want for them.:-)

The other thing that strikes me is that anyone using the oven has to know how to place the thermocouple, because everything depends on that.

My thermocoupler wire enters the box, from the side. The tip is bent down and rests on a clear spot, of one of the PCBs being cooked, in the middle of the oven.

Given the variety of people who may be using the oven in my situation, with varying levels of understanding about the process, I came up with this idea for locking down the heating profile, and eliminating operator error:

I would establish a total ON/OFF cycle time of, say, 5 seconds.  For each such cycle the average power level would be determined by the duty cycle.  The 5 seconds would be divided by 100, giving duty cycle granularity of 50 ms.  The present setting would be shown on a 2-digit display, and would be set using a rotary encoder.

All the tuning would be done by me ahead of time.  Using a TM-902C thermocouple/display and a sample board, I would experimentally develop a sequence of duty cycle settings that produces the proper curve for leaded paste, such as:

10 cycles at 50%
14 cycles at 70%
etc.

As I develop the sequence, it would be loaded into the "controller", which would then simply execute that sequence when you push the button.  It would be the same sequence, and the same heating profile, every time.   The user can follow the temperature with the TM-902C if he wants to, but he doesn't have to.  And the thermocouple is not even connected to the controller.  There would be a buzzer indicating when to open the door.

That would be a legitimate process, but will require a major amount of tweaking. And, you still won't overcome a lack of heat, or cooling speed. The reflow cycle curve is intended to raise the temperature slowly, in the beginning, to prevent thermal shock to the components. The heat then races to the solder reflow temp, stays there only long enough for the solder to flow. After that, the cooling process is the balance of getting the component out of the heat as quickly as possible, but not shock it, with cooling too fast.

I hope a somewhat longer period should work for pretty much anything.

See reflow cycle timing notes.

I'm also looking at the $8 Walmart Mainstays single-burner hotplate, with a steel circular saw blade on top.  In theory, it could be controlled by the same process.

You might use the type with clear glass lid, but the saw blade is a bad idea. You want the parts to be able to move and seat themselves. [Yes, they actually move.] And, the blade would be sinking massive amounts of heat.

I have considered an oven box and a heat gun.

PID Tuning instructions are available on the net and YouTube. It's not that difficult.
@ 7:21
http://www.pcbheaven.com/wikipages/PID_Theory/?p=1

Keep at it!

[Peabody]

I picked up one of the Walmart hot plates and a circular saw blade from Home Despot.  See attached pics.

The coils aren't quite level, so they touched the blade in some places, but not others, with resulting hot and cool spots.  So I replaced three screws with much longer ones so now the blade is about 1/2 inch above the coils.  I think that helps even out the heat somewhat, but at the cost of slower temperature changes.

However, the hot plate concentrates 1000W in a fairly small area.  The coils get red hot fairly quickly, and at a pretty low temperature setting the blade goes above 200C in a couple minutes.  Of course the temp control knob isn't usable in this setting because there's way too much hysteresis.  I have to have something like the saw blade for this hot plate.  Other models, like the Oster, have a built-in cast iron platters, but they have only one circle of the heating coil, whereas the Walmart has three at any point on the clock.

Based on playing with it a bit this afternoon, I think the hot plate and blade combination will have no trouble heating up fast enough, but it remains to be seen whether it can do that in the reflow stage without overshooting.  It may be that to prevent overshooting it would have to heat up more slowly than the protocol says should happen, and I guess then the question is which part of the protocol I should violate.

As for cooldown after reflow, I think the answer will be to put on oven mitts and simply remove the blade from the hot plate.  The blade will still be hot, but should cool off pretty quickly without the hot coils underneath, maybe even too quickly.

Anyway, I have everything I need in the junque box to breadboard this, with the exception of the relay.  But I found a US source of some new/old stock of Crydom 25A relays at a reasonable price, and assuming they are genuine, I think that should be enough.  Oh, and I'll also need a heat sink.

As for ovens, what I read online is that the bigger ovens with four elements and a fan are actually less responsive than the two-element no-fan ovens.  Not sure why that should be.

I will study the PID tuning notes.  Thanks for the link.

[brucehoult]

It will probably turn on and off once every 5 seconds for 4-5 minutes.  Think of it as a very slow PWM, with the duty cycle controlling the average power.  So probably less than 100 times per reflow.  But of course that's just one board, and someone may need to do several.  I don't really have any feel for how much it will be used, but every kit I see for this calls for an SSR.   I assume that's because switching 10 amps at other than zero crossing points would cause lots of noise, and wouldn't be so great for relay contacts.

I used $40-for-three Jaycar 433 MHz-controlled 240V relays like these...

https://www.jaycar.co.nz/remote-controlled-240v-mains-outlets/p/MS6140

... to control my Arduino-based home heating, switching a 2400W oil column heater every 30 seconds (or at least making a decision on whether to switch it every 30 seconds). This achieved normally around +/- 0.02 C temperature control in the living room&kitchen (as measured by a thermistor on the arduino, sampled 100 times a second then averaged with a 30 second time constant), with extreme variations of about 0.2 - 0.25 C when I turned the stove or electric kettle on or off.

The poor relays were doing this continuously from mid April to mid November, for about five years. I had one or two that got dodgy after a year, but one lasted three years.

I'm pretty sure they weren't designed for that number of switching events! Probably they expect a handful a day.

[SoundFan]

These chinese SSRs work well from my knowledge but their amperage is totally overrated (the 40A ones will switch maybe 15A with heatsink) so you should definitely get a 40A one to switch your 10A reflow oven.

[wraper]

These chinese SSRs work well from my knowledge but their amperage is totally overrated (the 40A ones will switch maybe 15A with heatsink) so you should definitely get a 40A one to switch your 10A reflow oven.

Whenever they work or not, they are not electrically safe and fireproof. If you look at some pictures of disassembled SSRs they simply don't have enough creepage distance between control and high voltage parts. And most likely you'll find dodgy optocoupler as well.

[Zero999]

If this project is ever completed, it will reside at a local makerspace.  With that, I would not feel comfortable trying to build an SSR from scratch.

No offence, but if you don't feel comfortable building the simple circuit posted by wraper, then you really shouldn't be attempting to build such a project.



It really isn't a complex circuit. You're switching 10A, so you need a higher current rating TRIAC than the BT-137, which can only handle 8A  For example the Q4015R5TP is rated to 15A and is widely available. At 10A, it will dissipate about 10W, so it will need a decent heat sink and thermal tab mounting kit. I'll help you with the thermal calculations for selecting the heat sink and mounting kit, if you ask. Earth the heats sink and keep the minimum distance between the mains and DC on the PCB above 5mm and it will be safe.

https://www2.mouser.com/ProductDetail/Littelfuse/Q4015R5TP?qs=sGAEpiMZZMvLfqGeKEshX8Jtl8j%252bpxKso7ZxoeqD3pz01KAeSOq5hw%3d%3d
https://www2.mouser.com/datasheet/2/240/Littelfuse_Thyristor_Qxx15xx_Qxx16xHx_Datasheet.pd-775524.pdf

[SoundFan]

Whenever they work or not, they are not electrically safe and fireproof. If you look at some pictures of disassembled SSRs they simply don't have enough creepage distance between control and high voltage parts. And most likely you'll find dodgy optocoupler as well.

Don't get me wrong, I never meant that they are risk-free (almost nothing that comes from China and controls mains voltage is) but as opposed to a lot of other Chinese stuff, these cheap fotek-clones have been proven to work well enough for the price as long as you don't trust their current rating. You can also add a cheap optocoupler IC to your circuit to protect your microcontroller (or whatever you're using for the project) a bit more.

[wraper]

You can also add a cheap optocoupler IC to your circuit to protect your microcontroller (or whatever you're using for the project) a bit more.

How this is supposed to work? You would need a separate isolated power supply to do this. Moreover, concern is not about killing microcontroller but possible electrocution.

On top of that if you are going to add optocoupler, throw out that crappy SSR and just use triac instead. For me it seems ridiculous fearing to make own SSR from a few parts and trusting unsafe counterfeit SSR instead.

[wraper]

How about having almost no isolation? If there happens to be a small solder blob over adjacent pads (and I'm pretty sure sometimes this happens), control input would be directly connected to mains.

[CatalinaWOW]

I don't doubt that you can build a safe SSR yourself.  Though the assurance that a new person will accomplish this isn't much higher than the safety of a random Ebay build.  There are many small safety gotchas involved.

What I question is that you can build one with a good enclosure (required for safety), appropriate input and output blocks, proper rated components from assured vendors and do it significantly cheaper than a good device from a solid distributor.  Even if you place zero value on your time. 

And also question whether the insurance company will believe your homebrew device met all safety standards, no matter how well done it is.  I can just see the courtroom testimony.  "Well I could have bought a proper unit.  Or I could have bought a unit that appeared to be proper, even though some people on line recommended against them.  Instead I trusted a guy I met on line and built something based on a partial schematic and a couple of design recommendations.  There is no way that the fire that started in this controlled had anything to do with my design."

The lawyers would have a field day.  Even though the design is sound, and even though there were no flaws in design or construction.

It may well be worth doing as a learning experience, or for the challenge.

[SoundFan]

How about having almost no isolation? If there happens to be a small solder blob over adjacent pads (and I'm pretty sure sometimes this happens), control input would be directly connected to mains.

That "pretty sure" is the problem. These SSR are used by hobbyists all over the world. If one can make his own or at least use something like an S102S02, it would be better of course. But most beginners are a bit scared of that, even though it is easy.