Dual voltage controller

[Steve Broady]

I think some testing is in order. Here’s the plan of attack:

Since I have a switch in series with the potentiometer, I don’t have to physically disconnect anything to isolate it for measurements, so that’s handy. I’ll start by measuring the value of the pot, more or less to satisfy my curiosity. Next, I’ll connect a load and turn the controller down as low as it’ll go (down from full, rather than up from zero), turn the power off, and measure the resistance. Do that for both voltages.

Once I have some accurate data, I can figure out what the values should be in order to get the most out of the system. I’ll calculate what resistances I need to put in parallel with the pot, then use the rotary switch to select the right one depending on voltage I’m running.

I did a quick search on Amazon for a dual potentiometer, but I didn’t see any with two distinct ranges. I’m sure they’re available, but I’m not sure if cost for such a specific part would make it word doing. My thought at the moment is to buy a selection of trim pots, then use ones with the appropriate ranges to fine tune the system. Presumably, once it’s done I won’t have to adjust it again.

[elbono]

I was thinking about a ganged pot too but needing two different values would probably make it a custom unit, very expensive if you could find it. Yummy's solution eliminates the need for different values.

I think you will find that the resistance needed for 120v is pretty close to half what is needed for 240v. If so Yummy's answer is the way to go.

The problem with a parallel resistor is the result won't be linear, centering the pot won't give you 1/2 resistance.

I like the resistance to be a little high so there is a small dead spot before it turns on. If it's a little low you can't turn it all the to zero volts.

Watch the power rating of these parts, all the specs I've seen call for a 2 watt potentiometer.

[Steve Broady]

Use one pot as you normally would on 240v . Connect one end of Both pots together. Put a switch between the Wipers ( centre terminals ) and switch on for 120v .leave the unused ends of the pots not connected .

I’m sorry, Yummyrum. I completely missed the point you were making here. That makes a lot of sense now.

Watch the power rating of these parts, all the specs I've seen call for a 2 watt potentiometer.

That’s one thing I haven’t seen any specs on, so I appreciate it. Does an SSR really need that big a resistor on the control side? I would have thought it was more or less a standard TTL circuit inside, running very low current. But I don’t know ouch about these things.

[elbono]

Watch the power rating of these parts, all the specs I've seen call for a 2 watt potentiometer.

That’s one thing I haven’t seen any specs on, so I appreciate it. Does an SSR really need that big a resistor on the control side? I would have thought it was more or less a standard TTL circuit inside, running very low current. But I don’t know ouch about these things.

I believe I've seen 2W several places but I know I see 470K-560K 2W in your pictures.

I believe I'll be running the still tomorrow, if I have time I'll measure the voltage on the pot.

The SSRs that you can drive with MCU's are opto isolated, all that's needed is turning on an led. I'm not sure about the internals of these guys but I have a sneaky suspicion the current the pot sees is related to the output load.

I put a 47k pot in series with my 470k pot for fine adjustment. It's not really required unless you're running a reflux column on the edge of flooding but it makes repeating previous power setting much less fidgety.

[Yummyrum]

Alas the internals of an SSRV is based on a 60+ year old circuit . There is no TTL .

The Main switching device is the Triac on the right . It switches when the voltage across the Diac connected to its control gate reaches the Diac breakdown voltage ( which is around 20-30v depending in device )

The Pot feeds current to charge the Capacitor on the left side of the Diac .
The less resistance , the quicker the capacitor charges and reaches the Diacs breakdown voltage .

Because the whole circuit is based around how long it takes to charge the capacitor to 20ish volts via the current through the pot , becomes the reason why the supply voltage has such an effect in the value of the Pot


Edit:At around 50% power , the voltage across the Pot is the highest (340vpeak) and the Pot is typically at around 200k .This is around 0.6W
But, as soon as the Triac fires, the voltage drops to almost zero , so at 50% power , the time the pot see's the maximum voltage is only 50% meaning the average dissipation in the Pot is only 0.3W

Although you can get away with using smaller wattage Pots , the internal physical construction means the dielectric insulation between the track, wiper and case is much smaller and more susceptible to electrical tracking and insulation brake-down .
I agree that using a larger 2 watt pot is a safer and more reliable option

[Yummyrum]

A bit off topic but as it was asked here, I thought I'd add this .

My trusty old controller that I was gifted from Panda just happened to die last week . I knew it had a tiny Pot in it but it's worked for 10 years or more so never bothered to replace it .
This things had a hard life , but as it turns out , the Pot failed .
The wiper looks perfect but the Tracks are not .

Hard to photograph but I suspect I mozzed myself when I was talking about tracking . The gap between the Aluminium shaft and the inner track is barely 0.25mm .There are tracking marks on the inner track and I suspect when it flashed over , the excess current fused the connections and the Pot tracks are open in several places .

Although not visible , a multimeter showed there were several places where the tracks were open circuit .

So there you go ,You can use a Small Pot , but eventually it will fail .

[shadylane]

I've done autopsies on pots that looked similar.
Not on a controller, but on old radio's, guitar amps and stuff like that.
Sometimes shit just happens, but I'd check the rest of the circuit, to be sure the pot didn't get murdered.

On a side note
2 watt pots are made better than smaller rated and cheaper ones.

[Yummyrum]

I've done autopsies on pots that looked similar.
Not on a controller, but on old radio's, guitar amps and stuff like that.
Sometimes shit just happens, but I'd check the rest of the circuit, to be sure the pot didn't get murdered.

On a side note
2 watt pots are made better than smaller rated and cheaper ones.

You got me worried about it being Murdered Shady . But when you mentioned Guitar amps you reminded me I had a bunch of stuff in the shed from when I made a Fender copy amp for my son . Sure enough , there was a B500K pot in the box of odds n sods.

Just fitted it and happy days .

[shadylane]

Not to talk bad about modern parts.
But much of the stuff made 50 years ago seems to be made tougher.
Ya might have to spend a couple minutes twisting the knob to clean the contacts.

The only electronic parts that time hurts, is electrolytic capacitors.
Even those can sometimes be convinced to come back to life without pissing oil and giving up the smoke.

[Yummyrum]

Hee hee , you’re preaching to the converted Shady
Did my apprenticeship as a TV technician and rode it to the end . Yup modern components suck . LOL , in the end customers didn’t want to spend anything and wanted tbe latest and greatest .

[shadylane]

Ya caught me editing while drunk Yummy.

[Yummyrum]

Cheers Shady

[shadylane]

Did my apprenticeship as a TV technician and rode it to the end . Yup modern components suck . LOL , in the end customers didn’t want to spend anything and wanted the latest and greatest .

TV techs have been a thing of the past for a long time.
Now everything is digital and unfixable. The old timer that was my ham radio Elmer used to make extra money repairing and aligning TV tuners as a part time hobby, while maintaining the Memphis, Tennessee TV station.

Harry Fairchild started out old school, as in home brewed one tube crystal-controlled transmitter and regenerative receiver days.

[Steve Broady]

This thing keeps changing on me. I’m really glad I used the bus bars now!

I took Yummyrum’s advice and used a double potentiometer (470k, 2W). They are in parallel when the selector is on the 120V side, and only one pot is being used on the 240V side. I also added a 20A receptacle on the back, so that I can plug in a hot plate or other 120V heat source.

While I was at it, I replaced most of the control wiring with some smaller gauge red wire, just to differentiate it from the power wiring. Electrons don’t care (at least, nobody has managed to get or to slow down long enough to ask if they have an opinion), but it helps me when looking at the wiring. I also had to solder the ground wire directly to the potentiometer case, since there was no ground lug. Easily done.

My plan is to strip some BadMo tribute rum tomorrow, so I’ll know soon enough if I did anything stupid. It works on 120V at least, using a heat gun as a test load. Hopefully I’m done making modifications. Just like I’m done modifying and building parts for the still. Riiiiight!

If anyone is interested, I can post a schematic of the finished product.

[venkman]

You guys are speaking my language here. I've built guitar effects units and repaired amps. I dig electronics. I was combing the controller threads looking for ideas and trying to understand the benefits/limitations between the "SCR" (TRIAC) circuits and the SSR/SSVR circuits people are using. I was thinking about adding a meter to my cheap eBay Chinese SCR to get a point of repeatability. After reading I was getting worried about the non-sinusoidal waveform frying the meter. Then I found this thread and it looks like Steve has built pretty much exactly the controller that would suit my needs (and then some.)

Very interested in seeing the schematic Steve Broady, if you have time to post it up. Thanks guys!

[Steve Broady]

Very interested in seeing the schematic Steve Broady, if you have time to post it up. Thanks guys!

I drew this up quickly this evening. I’m sure I could do a lot to make it neater, but hopefully it makes sense.

I forgot to note on the drawing that all the ground connections go to a third bus bar, since the case is plastic.

[Yummyrum]

Steve ,two things .

I noticed you drew the supply to the volt/ammeter is connected to the output side of the SSRV and not the input . This normally results in a burnt out meter .

Second and more of a worry . You have have switched one of the 240v hot wires but the other Hot wire is connected to your Neutral Busbar .
That means that when running on 240v , the Neutral pin on your 120V plug is live and potentially laying around on the floor for someone or something to touch .
( there would also be big Boomsky if you had both plugs plugged into their respective supplies at the same time ans switched on )

I think it would be better to use that Two pole change over switch to switch both Hot wires ( on 240) and both the Hot and Neutral ( on 120) to provide full isolation .
I know you are using that spare contact to switch the pot , but I think the safety is more important .

[shadylane]

Plus one Rummy for spotting hazards.

I'd use a male twistlock connector for the power into the box.
Then make Individual power cords for 120v and 240v that plug into it.

The switch would only need to connect the pots in parallel for 120v

[Steve Broady]

I’ll reply with my logic, not as an argument, but merely an explanation. I’m happy to learn, if I’m wrong.

you drew the supply to the volt/ammeter is connected to the output side of the SSRV and not the input . This normally results in a burnt out meter

I did that because I wasn’t sure if the output voltage would vary, or just the amperage. At least according to the meter, it varies down to about 3-5 volts before the meter shuts off.

What the cause for the meter failure on the output? Is there a way to protect the meter while still being able to read the output voltage? That seems like an important thing to know in order to determine the actual power being used.

Second and more of a worry . You have have switched one of the 240v hot wires but the other Hot wire is connected to your Neutral Busbar .
That means that when running on 240v , the Neutral pin on your 120V plug is live and potentially laying around on the floor for someone or something to touch .

You’re right. I isolated the hot pin on the 120V plug, but failed to isolate both. In truth, I’m so accustomed to the lower voltage wiring that I was thinking of the white wire as neutral, not hot.

That said, the way I designed this thing, the power cords can both be coiled and fit inside the top of the tool box. Only one cord should ever be in use at a time, while the other is safely stowed out of sight and out of reach. I agree, it’s not idiot proof, and it probably should be. I may look into adding a separate switch for the second pot, and use the contacts on the selector for full isolation. While I don’t see myself ever doing something that stupid, you never know what someone else might do when I’m not around.

[shadylane]

What the cause for the meter failure on the output? Is there a way to protect the meter while still being able to read the output voltage? That seems like an important thing to know in order to determine the actual power being used.

Don't know why the meters take a shit too early but here's my theory.
The meters are designed for measuring and running on a 60 cycle sin wave that slowly changes.
The output from a SSVR is a sawtooth wave form that shuts off instantly 120 times a second.
I figure that pounds the hell out of the meters cheap internal power supply.

[Yummyrum]

I'd use a male twistlock connector for the power into the box.
Then make Individual power cords for 120v and 240v that plug into it.

The switch would only need to connect the pots in parallel for 120v

Good work around Shady

I’ll reply with my logic, not as an argument, but merely an explanation. I’m happy to learn, if I’m wrong.
Yummyrum wrote: ↑13 Oct 2023, 08:08

you drew the supply to the volt/ammeter is connected to the output side of the SSRV and not the input . This normally results in a burnt out meter
I did that because I wasn’t sure if the output voltage would vary, or just the amperage. At least according to the meter, it varies down to about 3-5 volts before the meter shuts off.

What the cause for the meter failure on the output? Is there a way to protect the meter while still being able to read the output voltage? That seems like an important thing to know in order to determine the actual power being used.

I can't speak for all Volt/Ammeters, but the majority will last maybe a run or several then smoke up , especially when at lower voltages .
The reason is that they have a built in power supply to run the meter and display circuitry .The supply comes from the Voltage being measured and it is assumed that this will be a Sinewave as Shady said .
Alas , the output of the SSRV is not Sinusoidal and at lower voltages there seems to be massive spikey currents that flow in the components that are in the Meters Power supply and they burn out .
It's been a known problem here for many years .As such most just connect the Voltage input to the SSRV input and simply use the Current reading as the voltage reading will always show the Mains supply Voltage as you have ascertained .

Now , as you would know Power is Voltage x Current .So some meters that also display Power may or may not show the correct Power when the Voltage input is connected this way .
The reason is in how the meters derive the Power . If they simply Multiply the V and I readings ,then they will be wrong .
However , some better meters are sampling throughout the wave form and instantaneously multiplying the current and voltage readings and then
averaging them for the final power . This works because even if the meter measures the input voltage while there is zero current , V x I will equal 0 during the period of no conduction .

So I guess its a case of suck it and see. But be prepared to have to replace the meter .

Edit: Incidentally , there is a hack on You-tube somewhere where someone modifies one of these meters to remove the internal derived supply and adds an external one to mitigate the problem

[NormandieStill]

I’ll reply with my logic, not as an argument, but merely an explanation. I’m happy to learn, if I’m wrong.

...

While I don’t see myself ever doing something that stupid, you never know what someone else might do when I’m not around.

Forget "other people". While it's good practice to ensure that your devices are safe in the face of the wandering idiot. Don't fool yourself into thinking that you are not also a wandering idiot. Most of my near misses have occurred when I was trying to get something done too fast or with too much going on in my head. These range from blowing a chip out of a table whole defrosting a freezer (slid it into the control box... cheap and shitty fridge / freezer) to blowing out the inlet water tube on a condenser because I was thinking through how to make it safer / more reliable.

[elbono]

I believe I've seen 2W several places but I know I see 470K-560K 2W in your pictures.

I believe I'll be running the still tomorrow, if I have time I'll measure the voltage on the pot.

I finally finished some marathon spirit runs and made some measurements.

We are using a potentiometer as a rheostat so that's really a current rating. 470k ohms at 2W comes out to about 2 milliamps.

I got:
Full CCW - 0.5 mA
1/4 CW - 0.6 mA
1/2 CW - 0.8 mA
3/4 CW - 1.2 mA

So I dont think a pot less than 2w is advisable.

[shadylane]

I finally finished some marathon spirit runs and made some measurements.

We are using a potentiometer as a rheostat so that's really a current rating. 470k ohms at 2W comes out to about 2 milliamps.

I got:
Full CCW - 0.5 mA
1/4 CW - 0.6 mA
1/2 CW - 0.8 mA
3/4 CW - 1.2 mA

So I dont think a pot less than 2w is advisable.

Thanks for taking the time and sharing the info with us.

[elbono]

Thanks for taking the time and sharing the info with us.

Thanks, I'm hear to learn. If I have a question I want the answer. I believe the answer is best used when shared.

[Steve Broady]

Thanks, I'm hear to learn. If I have a question I want the answer. I believe the answer is best used when shared.

Agreed! And I thank you for both pointing out the rating on the pot and then testing the importance of that. You stopped me from buying some severely underrated components. Thank you.

[shadylane]

I'd use a male twistlock connector for the power into the box.
Then make Individual power cords for 120v and 240v that plug into it.

Good work around Shady

Not a work around, I think it's the best way to start.
I figure, when building something for bragging rights, do it right.

Something like this.

[Yummyrum]

I'd use a male twistlock connector for the power into the box.
Then make Individual power cords for 120v and 240v that plug into it.

Good work around Shady

Not a work around, I think it's the best way to start.
I figure, when building something for bragging rights, do it right.

Something like this.

On second thoughts , agree with you 100% Shady .

[Steve Broady]

Time for an update. I took some of the advice here to heart, and started over with v2.0

The most important change is that I went from having two power cords to a single NEMA L6-30 plug, with separate cords for both 120V and 240V circuits. I also replaced the L5-30 plug and socket on the element (which I had used because that’s the extension cord that I found cheap on Amazon, and didn’t know better at the time) with L6-30. Because standards.

The new meter panel is a slightly modified D69-2049. I separated the power and voltage sensing circuits, so that it always gets full voltage but can measure the output voltage.

The other obvious change is that I replaced the rotary switch with a pair of toggle switches. Both are in the SSR control circuit. One cuts the circuit completely, turning off the output while leaving the fan and meter powered. The other puts a second potentiometer in series, extending the control range when running on 120V.

The wiring inside is a little bit of a mess. I may tidy it up one of these days, if I ever get the motivation. Everything is nice and solid, run on heavy duty bus bars, solid copper crimped terminals, etc. I’m as confident as I can be in the quality of the work, neatness notwithstanding.

I even have room inside the larger case to store both power cords. I don’t see any reason why I can’t have them in there while running, since they’re well away from the heat sink and it doesn’t seem to get that hot anyway.

Finally, I was able to use this to test a hot plate which I bought to run a small still. I wasn’t sure if the control on the hot plate was varying the power, or just cycling it via a thermostat. Simple way to find out. I plugged it into my control box and read the power as I adjusted the controls. 1400W across the board. Good to know!

[Yummyrum]

Nice re-build Steve .

Just a silly question , but is that middle wire on the Pot soldered ?
It just looks like you looped the wire through the eyelet but its not soldered

IMG_9144.jpeg (18.98 KiB) Viewed 1507 times