Brandon Bearden Posted October 11, 2022 Author Share Posted October 11, 2022 I forgot about Allied, I used to use them too. Mostly I just look at Mouser and Digikey. However, I am in luck. I was able to find a new E5AC-TQX4ASM-080 on ebay shipped from China. I called Omron tech support again, those guys are great btw, and they told me that was indeed a euro/asia market model but it was identical in every way to the USA version. He confirmed that the 060 models are equivalent to the 000 models in the USA and that 080 are equivalent to the 019 models. So, I just ordered it. I am going to keep the one at digikey on order in case the tech support guy was mistaken about the interface or features, also I may just get sand when it arrives. If it all checks out, then I will return the 000 one to digikey and cancel the 019 order. I am stoked! I was not at all looking forward to working without those inputs, now I have 6 I am going to work on the wiring diagram and know who to run it by when I am done. I will call out all the component choices but don't expect anyone to do a full review with spec sheets and all. I have done a bit of this kind of work in the past when I was developing an open source programable power strip and control module for reef aquariums. Basically, I took the Arduino Mega board and had an EE take all of my additional component designs and build a full new board and I bought the SMD soldering equipment and built a bunch of boards. Its purpose was to monitor PH, DO, temp and bunch of other parameters, switch a power strip, and serve all that up over the web while communicating to a web server. So, it had on board wifi and such. I was able to even code up an event based mini OS for it with interrupts and such. The code was my safe zone as that is what I really do. The EE was huge learning journey. It was a pretty massive project and had great promise, but my new wife insisted I was taking too much time so I abandoned it. That was about 9 or 10 years ago. That was my only real foray into the world of embedded electronics. Anywho, I ramble. I just wanted to thank you again for the advice an input. More to come soon! Link to comment Share on other sites More sharing options...
Brian Dougherty Posted October 11, 2022 Share Posted October 11, 2022 I'm sure you'll get it all figured out. Happy to look at your schematic when you get to it. It'll be interesting to see it run when you're done. It should be quite a beast! -Brian Link to comment Share on other sites More sharing options...
Brandon Bearden Posted October 12, 2022 Author Share Posted October 12, 2022 @Brian Dougherty I am having trouble understanding the pinout on this wiring diagram for the event inputs. If I were to wire a switch to an event input, say input 1, how would I do that here? And then if I wanted to use event input 2? It confusing they are sharing pins. Same for the Aux Outputs... Link to comment Share on other sites More sharing options...
Brian Dougherty Posted October 12, 2022 Share Posted October 12, 2022 Well, based on nothing more than the pictures you posted, I would say pin 16 is the common for event input 1 and 2. If you were going to use EV1, wire the switch between pins 16 and 17. EV2 would require a switch between pin 16 and 17. If you use both, pin 16 will have two connections. Proceed cautiously since I'm basing that on the pics, and not really researching it properly. What is probably going on inside is that each input has an internal pull-up resistor tied to the supply rail to take the input to a logic high level when nothing is connected. By connecting that input pin to the common pin denoted with the (-) symbol, you can impart a logic low on the input. Don't get confused by the notion of an input being on when it is presented with a logic low. This is pretty typical because it simplifies the internal electronics somewhat. -Brian Link to comment Share on other sites More sharing options...
Brandon Bearden Posted October 12, 2022 Author Share Posted October 12, 2022 Thanks Brian. I found this in a different document. You are correct. One other question - I hope this is the last one. The datasheet states: "Separate input leads and power lines in order to prevent external noise." I keep trying to think of what this means, but I don't know. Maybe this helps to answer. I didn't realize this thing would be so darn sensitive. They also have this fun bit of wording under their precautions: They also suggest warming up the controller for 30 minutes prior to use, which is kind of crazy because part of the point of having a high powered system is to have less wait time for heats. I also see that the system doesn't want heaters over 50A, especially if I am going to use the current transformer to monitor for shorts or burnout. I am going to tune the resistance wire length to ensure I stay just under 12,000 watts, maybe target 11,500. Link to comment Share on other sites More sharing options...
Brian Dougherty Posted October 12, 2022 Share Posted October 12, 2022 (edited) In general, you always want to keep high power wires separated from low voltage signal wires. In this particular case, you will want to avoid running the heater element wires in parallel or in close proximity to the thermocouple wires. Everything else can probably be pretty close to the power wires. The current pulses in the heater wires can easily cause noise in the thermocouple wires that would be significant relative to the size of the very small signals that come from the thermocouple in the first place. This noise may cause the controller to "Read" the wrong temperature value. The closer the wires are, and the more distance they run in parallel, the more noise you will get. I have spent days chasing problems in control systems cause by this sort of thing. The 30 minute warm up time is only to get the ultimate accuracy. Electronics drift a bit with temperature, so that is just to get all the components to steady state so the values stop drifting. I think you can safely ignore that for a heat treat oven. Not sure why the 50A limit is there, but it may be caused by a voltage limit in the component that reads the CT signals. The signal out of the CT will be proportional to the current. It is possible at over 50A,the signal would damage the amplifier it is being fed to. Edited October 12, 2022 by Brian Dougherty -Brian Link to comment Share on other sites More sharing options...
Brandon Bearden Posted October 13, 2022 Author Share Posted October 13, 2022 I am considering a zero-cross DPST relay instead of two SPST relay's. Anyone have thoughts on any issues I may encounter with a zero-cross relay? I know what it does, but with two poles of 120 each, my knowledge of the way electricity works in a home wiring is not robust enough to understand if this will create any unintended consequences. I also just noticed that besides auto tuning, this controller also has robust tuning that is designed for system stability. This will be fun to play with if I can finalize all the other component selections... always an extremely time consuming process. There are so many good components to choose, but then availability kicks in and kicks you in the butt. Link to comment Share on other sites More sharing options...
Brian Dougherty Posted October 13, 2022 Share Posted October 13, 2022 Can you put up a link to the datasheet? Since this is a zero-crossing relay, I am assuming it is an SSR, in which case, this is probably its intended use. There are a few folks on here that know a lot more about house wiring than I do so I would prefer to let them chime in, but essentially the two 120V poles you see here in the US are fake. What you really have is 240V across the two lines they run to the house. if you look at each of those relative to ground you see half the voltage. -Brian Link to comment Share on other sites More sharing options...
Brandon Bearden Posted October 13, 2022 Author Share Posted October 13, 2022 Sure, here is the datasheet. It says it is for AC-51 loads which apparently is "Induction-free or slightly inductive loads, resistance furnaces" My only real concern was that since it is zero crossing, is there a time when the two legs of the 240v, which I know is split at the breaker box into two 120 legs that then come back together for the 240, are not in phase? Meaning, could the wave sin be 180 off, so they are opposite which would never trip the zero cross threshold? One other minor concern was the added switching time, because I don't know how many pulses per second something like this normally runs. My only work with a PWM type system is with small DC components. I am sure this is not really PWM, but in principle, it is very much like that. So, if this thing wanted to pulse 10 times a second, at 60 hertz, with the possible lag in switch time, maybe that is too much and it will only get to say 5 per second or something. Again, this is me not knowing how the controller will really work to create the duty cycle. Link to comment Share on other sites More sharing options...
Brian Dougherty Posted October 13, 2022 Share Posted October 13, 2022 I've never used one of those so take this with a bit of salt, but it looks to me as if it should work. It is clearly designed to be use to control the power in a 3-phase load so they must have designed the zero crossing detection to work accordingly. The only thing you are doing that is a bit odd with respect to the intended usage is that you are not driving a 3 phase load. However, I don't think that is an issue. (Someone chime in here if you think I'm wrong) The reason for using zero-crossing relays is so that you don't hit a cold heater element with the peak of the sign wave. This is a pretty standard approach in heater control systems, so I would assume the controller is chopping at a frequency significantly slower than 60hz. -Brian Link to comment Share on other sites More sharing options...
Brandon Bearden Posted October 13, 2022 Author Share Posted October 13, 2022 (edited) Good point about the 3 phase detection. I was finally able to find a 2 phase model from the same company. They were marked SPST on DigiKey which is why I didn't see them at first. I am going with SOB967660 and a huge fan driven heatsink for it made by the same company. I have every component selected now except the lid switch and the circuit protection equipment. Edited October 13, 2022 by Brandon Bearden Link to comment Share on other sites More sharing options...
Dan Hertzson Posted October 13, 2022 Share Posted October 13, 2022 If accuracy is important to you, you should consider using thermocouple wire between the thermocouple and controller, not normal copper wire. Link to comment Share on other sites More sharing options...
Brandon Bearden Posted October 13, 2022 Author Share Posted October 13, 2022 (edited) @Dan Hertzson Interesting. I assumed that the leads off a thermocouple could be extended. I was trying shield the wire from interference. The leads on the ones I have are long enough to reach without any extension, maybe I can shield them another way, or maybe it is not needed at all. The leads on the one I have from Omega, it is not a cheap ($90), looks like regular tinned wire - perhaps it is not. Edited October 13, 2022 by Brandon Bearden grammar Link to comment Share on other sites More sharing options...
Brandon Bearden Posted October 13, 2022 Author Share Posted October 13, 2022 Final wiring diagram unless someone shows me something wrong and I need to update. Link to comment Share on other sites More sharing options...
Brian Dougherty Posted October 13, 2022 Share Posted October 13, 2022 A few minor comments: Generally the L1/L2 pins are the input side of the SSR and the other pins would be the load side. You might want to double check to make sure that the controller actually supplies the drive voltage for the SSR, and that voltage matches the needs of the SSR. I've seen a number of controllers that require either an external supply, or that you had to jumper over to an on board supply. (I suspect you have it right, but I've been bitten by this same thing in the past) For your current transformer signal wires, you probably mean shielded twisted pair. The twisted pair is more important than the shield. Also, if you use shielded wire, only connect the shield drain at one end of the wire. (I'd ground the drain at the controller end) Don't worry about shielding the drive lines for the SSR or the power supply to the controller. Doing so might actually make things noisier. Also to Dan's comment above about the TC wire: Thermocouples are two dissimilar metals connected together. They produce a current proportional to the temperature. The lead wires of the thermocouples are made up of the two different metals so if you connect them to copper wire, you create another thermocouple at that connection. The end result is that the instrumentation will not only see the signal of thermocouple in the oven, but also of the other two you created by connecting extension leads. There are ways to do this so that they are canceled out, and that is usually what is done inside the instrumentation so you can safely connect the TC wires to the terminal blocks. -Brian Link to comment Share on other sites More sharing options...
Brandon Bearden Posted October 13, 2022 Author Share Posted October 13, 2022 Thanks for taking al look, Brian. In order from above. Ah, I thought L1 meant load and I thought the diagram on the datasheet concurred with that, but I will double check. Yes, it works out, the controller is 12v and the SSR is 10-30v. But, yeah, good to check. Yes, I do mean twisted pair. Good note on the ground one side bit. Good to know about not using that wire for the drive lines or power supply. I figured that would be a good thing. I did not know that about thermocouples. Now I do. Thank you both. Link to comment Share on other sites More sharing options...
Brian Dougherty Posted October 13, 2022 Share Posted October 13, 2022 3 minutes ago, Brandon Bearden said: Ah, I thought L1 meant load and I thought the diagram on the datasheet concurred with that, but I will double check. Understandable, but I'm pretty sure you'll find that is the Line side, not the load side. -Brian Link to comment Share on other sites More sharing options...
Brandon Bearden Posted October 13, 2022 Author Share Posted October 13, 2022 lol. Double L. thanks. Link to comment Share on other sites More sharing options...
Brian Dougherty Posted October 13, 2022 Share Posted October 13, 2022 No worries Thinking about it a bit more, you have multiple circuit breakers. That generally isn't done, especially since they are all 60A. Typical practice would be to put one on the input side of your device. It also seems a bit wrong to have it be the same rating as the house circuit protection limit. I would want to shoot for a little lower so your device will trip before the house power. The circuit breaker in the house is there to protect the house wiring, not the device. Ideally the device won't challenge the house. I think you were going to run <50A? You might shoot for protection closer to that limit. I am not really experienced with over current protection design, so this would be a good place to get another opinion. -Brian Link to comment Share on other sites More sharing options...
Brandon Bearden Posted October 13, 2022 Author Share Posted October 13, 2022 (edited) The first one is in the breaker panel coming off to the wall outlet. The second one is so that I have a disconnect and circuit protection in case of short in the unit. The third one is because the spec sheets for the relay say to protect the relay with either a fuse or a breaker from the resistive load from shorts. I know that the enclosure breaker should trip very quickly, but the company explains that the time it takes to wreck the electronics inside the SSR is so short, that overload/short protection on the other side of the load is too slow. I could put a fuse in instead, I just like breakers. I thought about closer to 50 amps too, but I think that violates some sort of load rule for breakers. Thoughts? Edited October 13, 2022 by Brandon Bearden Link to comment Share on other sites More sharing options...
Brian Dougherty Posted October 13, 2022 Share Posted October 13, 2022 I'm not sure. That's gotten out of my element. Electrically it will work, but I don't know what best practice is. Maybe someone else will chime in with a more experienced opinion -Brian Link to comment Share on other sites More sharing options...
Brandon Bearden Posted October 14, 2022 Author Share Posted October 14, 2022 I have to thank you again, Brian. It is joy to learn, seriously. I always learn so much in projects like this and from reading datasheets - which is art. So, I think that the amp rating doesn't matter so much in this case. In-rush currents can be much higher than load currents. Also, the peak current for the SSR I chose is 75A where the breaker current max is 60A. Really, what matters more is the time. Apparently, the specs on the relay state that 10ms is the time limit on the connection break. A fast blow fuse could accomplish this with proper selection. Apparently most breakers can't react fast enough. I found this great pdf talking about SSR protection with fuses. The bottom line, it is not so simple to design. I am glad you brought this up because otherwise, I may well have had that breaker sitting there doing nothing with me thinking the relay was protected. They don't publish the l2t rating of that particular breaker, but I bet it is not 6000 or less, which is what is needed in this case. I have actually reached out the manufacture of the relay and they are going to help me figure out exactly what breaker or fuse that is available in the market I should use. Thank goodness I don't have to figure it out Also, derating, I need to be sure the enclosure of the oven doesn't get hotter than 100f or about 37.5c. Even at 112f that breaker derates to 52.2A. Link to comment Share on other sites More sharing options...
Dan Hertzson Posted October 14, 2022 Share Posted October 14, 2022 16 hours ago, Brian Dougherty said: Also to Dan's comment above about the TC wire: Thermocouples are two dissimilar metals connected together. They produce a current proportional to the temperature. The lead wires of the thermocouples are made up of the two different metals so if you connect them to copper wire, you create another thermocouple at that connection. The end result is that the instrumentation will not only see the signal of thermocouple in the oven, but also of the other two you created by connecting extension leads. There are ways to do this so that they are canceled out, and that is usually what is done inside the instrumentation so you can safely connect the TC wires to the terminal blocks Also worth noting that the two wires from your thermocouple, if properly selected, need to be connected to the correct terminals on the controller so the compensation that Brian notes is effective. You probably are aware of this already, but worth mentioning. The wires should be labeled. Link to comment Share on other sites More sharing options...
Joël Mercier Posted October 14, 2022 Share Posted October 14, 2022 1 hour ago, Dan Hertzson said: copper wire Let's say at an average aus temp like 1475°f, how far off would copper wires give you false readings? Are we talking 1-2° or 10-20°? Link to comment Share on other sites More sharing options...
Brandon Bearden Posted October 14, 2022 Author Share Posted October 14, 2022 (edited) So, I am trying to decide what to compromise on here. I talked to the Kanthal rep and they said that APM wire will last 2-3 times longer in service than A-1 wire. With this oven design, I am going to try to figure out how build it so that I can easily replace the element, however, that may just not be feasible with the baffle. Also, I don't want to do this twice. It is dang hard to find APM wire anywhere. I found a place that has 3.5mm APM wire. The problem is, that to stay within spec on the surface area wattage I have to drop the oven to 10,900 watts and that maxes the SA spec out. If I push beyond the recommended spec to 3.27 w/cm^3, (3 or lower is the recommended spec), then I can have 11,945 watts. Kanthal will sell me 100 lbs of the stuff when I only need 22 lol. Anyone need 78 lbs of 4mm APM wire four months from now? OR... I could just abandon the APM wire and go with A-1 wire using 6 awg and that would output 11,945 with a SA value of 2.01 which is awesome. Also, of course cheaper. Any suggestions for me? On another note. This oven is going to weigh like 200 lbs Edited October 14, 2022 by Brandon Bearden Did I mention I tend to go overkill if that wasn't already obvious? Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now