0

I'd like to run a kiln on DC power. I make colored glass and the AC power seems to be affecting the chemistry of the glass. It sounds strange but it's been a long process of eliminating a variety of possibilities.

How do you connect a bridge rectifier to shop power (US 240V) and how would you make connections to the heating elements?

I usually just run L1 and L2 to either end of the elements so I'm guessing it would just use the + and - of the DC terminals.

My shop doesn't have a neutral so would I just use my ground and L1 as the inputs. Does L2 just get capped off? I realize I'm losing a lot of voltage this way but the kiln is only pulling about 100 volts at 50 amperes, anyway. It's mostly just to test and see if the chemistry works better.

JRE
  • 67,678
  • 8
  • 104
  • 179
Daniel Van Antwerp
  • 69
  • 5
  • 1
    _"I realize I'm **losing a lot of voltage** this way but the kiln is only pulling about 100 volts at 50 amps"_ - No, you will only lose about 1V across the rectifier. However if you read the 'DC' voltage with a multimeter it will show ~10% less because the meter reads average not rms on DC. – Bruce Abbott Jul 21 '22 at 02:31
  • 1
    A heater does not pull voltage! At about 100 V your killn pulls 50 A is that right? – Uwe Jul 21 '22 at 05:58
  • 1
    Do you mean you're trying to run a 100V kiln off a 240V supply? Using ground as a bootleg neutral is dangerous. Find a proper solution. – Simon B Jul 21 '22 at 09:56
  • You should know that you won't get **dc** just by using a bridge rectifier...you will get rectified ac which is not the same. – Elliot Alderson Jul 21 '22 at 10:45
  • My terminology was bad. The kiln runs using a proportional control SCR. At 2100F it is running a steady 100V at 48 amps. The kiln is made for running 240V and is on a 60 amp service breaker. – Daniel Van Antwerp Jul 21 '22 at 13:26
  • 2
    Please elaborate on _”I make colored glass and the AC power seems to be affecting the chemistry of the glass.”_ If the glass does not come into contact with the heater, this makes no sense. – winny Jul 21 '22 at 20:49
  • What is the quality affected? Are these birefringence colors? – Tony Stewart EE75 Jul 21 '22 at 21:12
  • These are colors made with either copper or silver. Under the proper conditions, silver becomes colloidal and is called silver opal, or chalcedony. Copper can aggregate and gives copper ruby. Also, I mentioned a MnO2 purple glass that requires oxidizing conditions but is very faint in this furnace. The MnO2 result was very telling to me that the redox chemistry is somehow affected. – Daniel Van Antwerp Jul 21 '22 at 21:42
  • @winny I apologize for the poor wording. I was trying to ask the question without getting into the weeds of my problem. It's very confusing and I understand the skepticism I am getting. But, AC current does produce an electric field when passing through a wound coil does it not? Also, if you gather glass from this furnace and ground yourself it will shock you. Molten glass can carry a charge and while the elements aren't in contact with the glass it is crossing through air like induction. I built this kiln and there is very little space between pots and elements but they don't touch. – Daniel Van Antwerp Jul 21 '22 at 21:50
  • _”Also, if you gather glass from this furnace and ground yourself it will shock you.”_ If you are making contact between the kiln and ground and you get a shock, your kiln is leaking current and your RCB should trip. Is it a buzzing tingling sensation? _”Molten glass can carry a charge”_ Glass is a good insulator but a blob of glass has nC levels of charge at best. _”glass it is crossing through air like induction”_ Air does not create induction. If you do induce anything in the glass once out from the kiln, it’s uV at best, if not less. – winny Jul 21 '22 at 22:03
  • I realize you've moved on to troubleshooting other issues by now, but is it possible that the chemistry was not affected by electric fields, but rather the vibrations in the heating elements caused by SCR switching transients? (If so, changing to DC will help.) – Theodore Sep 08 '22 at 19:47
  • @Theodore The change to DC is still ongoing. I'm hopeful it will make a difference no matter the reason. The difficulty in building parts that can withstand these conditions was hard to predict. I'm bumping my way through it with some excellent and invaluable help from members of this resource. – Daniel Van Antwerp Sep 08 '22 at 19:57

5 Answers5

3

Yes, you'd run AC L1 and L2 to the rectifier "~" terminals and the kiln elements to "+" and "-". You do not need (or want) to connect neutral. And, do not connect the heater element to safety ground: that is unsafe!

(I'm not sure what you mean by "kiln is only pulling 100 volts". Do you mean watts? And it would be up to 12kW if it's on a 50A circuit. MORE: ok, there is an SCR chopping the AC to a bit less than 42% of full power. This could interact with the rectifier you propose, but probably will still work.)

Here's a related question about SCRs controlling heaters: How to use SCR to control the power of a heater. I suggest you study this and understand better how your controller works before you try this.

Say you wire up the rectifier. It will not be putting out smooth DC. Instead, it will be pulses at 2x the line rate, and when applied to the heater it will do practically the same thing as AC. That is, elements will emit heat that has a small 2x line-rate ripple element to it. The heating element ends will be swinging from 0 to +165V (+) and 0 to -165V (-), respectively, vs. +/-165V for both ends when using straight AC.

Here's a sim showing that (simulate it here)

enter image description here

Given more time we could simulate it with the SCR chop.

You could smooth the ripple by adding capacitors. Then you will be left with having a DC bias across the element. Then this can test your AC power hypothesis. I'd be leery of this however, as this will interfere with the SCR and possibly damage it.

Which brings up another hypothesis: ionization. AC or DC, the elements will have a voltage gradient across them. As stated, the ends will be swinging +/-165V with the 0V point roughly in the middle. In heated air these differences will create ionization paths, which could be influencing your melt.

A possible experiment: add a grounded screen of high-temp metal (Iconel or such like) between the elements the crucible. This would block the ionization path near the crucible.

hacktastical
  • 49,832
  • 2
  • 47
  • 138
  • 1
    Not entirely. A FWB is enough to charge a battery, after all, if crudely. Presumably, it will suffice to eliminate whatever OP is after, as well as causing electrolysis and electromigration. – Tim Williams Jul 21 '22 at 02:14
  • 1
    One thing that happens is that there is a voltage gradient across the element, that changes polarity with AC. With smooth DC the gradient will be constant. Perhaps try a mica sheet over the element to block ionization? – hacktastical Jul 21 '22 at 02:44
  • Mica melts at about the same temp as glass... about 1300C. So maybe not possible to use that. – hacktastical Jul 21 '22 at 20:21
  • How does the addition of capacitors affect the SCR? I am installing them after the bridge rectifier which also sits downstream of the SCR. If it just outputs AC current based on what the controller tells it, then how are capacitors influencing this? – Daniel Van Antwerp Jul 25 '22 at 21:45
  • Adding capacitors would alter the I/V relationship that might mess with the SCR triggering, vs. the purely resistive load of the heater element. It’s similar to the problem of using a dimmer on a fluorescent ballast, except that has inductance. – hacktastical Jul 26 '22 at 00:32
2

Using ground as neutral is a definite no-no.

Here's the schematic.

enter image description here

The bridge rectifier is to be mounted on an appropriate heat sink. The heat dissipated would be about 100 W.

Silicone heat sink paste is to be applied between the rectifier and the heat sink.

The output RMS voltage would be 240 V minus the diode forward voltage drop of 2 V.

vu2nan
  • 15,929
  • 1
  • 14
  • 42
  • But the DC voltage of the bridge output is too high for the killn with 100 V and 50 A – Uwe Jul 21 '22 at 06:46
  • 1
    @Uwe - Hi, 'I usually just run L1 and L2 to either end of the elements' & 'but the kiln is only pulling about 100 volts at 50 amperes' are contradictory statements and I have presumed that the first one is correct. – vu2nan Jul 21 '22 at 11:04
  • Thank you for clarifying the install. I read a post about using 2 bridge rectifiers on a US 240V system and became confused by the way the installation was discussed. It seemed to say you could only run the rectifier off a single leg, but I'm really glad to read the consensus is to use both legs. – Daniel Van Antwerp Jul 21 '22 at 13:30
  • Hi Daniel, Please clarify whether the heating element is rated for 240 V or 120 V. – vu2nan Jul 21 '22 at 13:44
  • Hi, This is a 240V system. The heating elements are 11g kanthal A1. It has 4 element rows running around a 4-sided interior. Each side is 15" long. Inside the kiln sit 2 glass crucibles that can hold about 10lbs of glass. There are 2 parallel element paths in which 2 elements are run in series. Total amps are 48. – Daniel Van Antwerp Jul 21 '22 at 13:51
  • Hi Daniel, Got it! Thank you for furnishing the details. It's clear now that the heater rating is 240 V and 100 V 50 A pertains only to the proportional controller output. I am sorry I missed accepting your thanks for my answer. You are most welcome! – vu2nan Jul 21 '22 at 14:35
1

Yeah, that will probably do. Connect L1 and L2 to "~" and elements to +/-. The same power will flow (minus a couple volts), and the same controller (thyristor / phase control?) will do.

Note that fuses do not function on DC, or rather, special types are required to break DC. If the element is fused directly, move the fuses ahead of the rectifier, on the AC side. They will work normally there. Probably, the fuses are in the controller, and this is no problem at all.

Note also the rectifier requires heatsinking. Mount it on a heatsink capable of dissipating about 100W. Shop for this, or a temp rise of 1°C/W or less. Drill and tap holes in the heatsink (if not already on centers for the device), and use thermal grease in the joint between device and heatsink. Place the heatsink in an area with free, cool (ambient) airflow. Place guards over the connections, and probably connect an electrical grounding lug to the heatsink for safety.

I doubt that you will find a difference; I don't know of any chemical effects that AC might cause, and the induced fields are also small. More likely you need control of ambient gases, including oxygen balance (O2 vs. CO2 vs. CO and H2; inert gas purging?), or trace gases like H2S, SO2 and NOx. Note that these can adsorb into the porous kiln walls, so an extensive purge cycle may be necessary, if you choose to use inert gasses.

Tim Williams
  • 22,874
  • 1
  • 20
  • 71
  • I appreciate your response. There has been considerable trial and lots of error! I am trying to achieve reduced oxygen condition. Most glass color chemists find this to be difficult using electric kilns, preferring gas and a rich flame. Even in a gas kiln, reducing agents need to be added. What I am finding is that the reduction state is already too high. I can't control it or predict how to counteract it. My hope is to change a "condition" caused by a bunch of coiled coils running AC power. Can ripple DC current cause the same induction-like effects that pure AC can? – Daniel Van Antwerp Jul 21 '22 at 13:42
  • @DanielVanAntwerp You know it's just air in there, right? There's no combustion, it's just oxygen and nitrogen, hot. So of course it must be oxidizing. You have to add a fuel, a reducing agent, to consume the oxygen. Your underlying problem is chemical, not electrical. I know a bit of chemistry myself, but you may find richer discussion on the arts&crafts or chemistry stacks. – Tim Williams Jul 21 '22 at 18:30
  • I've had my reducing formulas working in another electric kiln. Tin oxide, sugars are tricks of the trade. I can even frit the glass once to remove O2. Then I made this smaller one and everything went haywire. For instance, MnO2 which requires oxidation, takes 50% more than it should to make a deep purple. Electrolysis pushes electron transfer which is otherwise chemically unfavorable. I cannot guess the actual complexities as to what is going on, but something is amiss. It's well known that molten glass conducts electricity pretty well. DC may be no better than AC, but it's only $30 to try. – Daniel Van Antwerp Jul 21 '22 at 19:11
1

You're on a wild goose chase. The change from AC to DC will not affect the chemistry. You have not shown this at all, and the "it seems" part is pure conjecture.

Yeah, it's not impossible that there's some ferrochemical stuff going on, but I place it so low on the list of possibilities that unless it were a well documented issue in the glass industry, with publications about it, you'd best just forget about that.

You haven't compared identical kilns, one running at DC, another at AC. Yes, they'd need to be otherwise identical, and you would modify one for DC. And you'd need to instrument them with a few thermocouples to make sure they really generate the same temperatures inside.

You're wasting your time with this here. You should be asking on chemistry stack exchange about the chemistry of this, and how you determined what your problem is. Bet you dollars to donuts that the problem is in the chemistry and how you reason about or measure things there.

Given the power levels involved, and the electrical safety, any suggestions given in other answers on how to modify your kiln are entirely inappropriate and dangerous. It is really a case "if you have to ask, you shouldn't be doing it". Just because glass is more dangerous doesn't mean the electrical hazards are something to scoff at. They are in addition to hazards from the hot/stressed glass. I don't believe there is any chance that you can execute this change safely, and it's not necessary anyway.

Kuba hasn't forgotten Monica
  • 32,734
  • 1
  • 38
  • 103
  • I, like you, doubt that the AC current in the heating element is going to explain whatever objectionable phenomenon the OP is seeing. But it seems that you wrote an "answer" which actually does not answer the question at all. Also, even though the OP may not be qualified to do this safely, it does not follow that nobody else is qualified to do this safely. There could be some value in answering this question for people who come along later after searching on a search engine. – user57037 Jul 21 '22 at 21:24
  • We'll see. I have a PhD in chemistry and I've been making glass batch for 8 years or so. Been blowing glass for about 20 years. I've eliminated a lot of the possibilities and this is just something I need to try. As a scientist in his 50s, I've learned not to shrug ideas off so easily. Glass chemists are very secretive. There are only a few books on the subject and they are pretty old. Most of it's done in gas equipment and only recently have coiled electric melters become popular. Not many people use them to make color...hmmm, I wonder why that might be? Perhaps if it works I'll publish. – Daniel Van Antwerp Jul 21 '22 at 21:33
  • @DanielVanAntwerp Ah, impressive! Then, do you have any means to isolate, correlate and analyze? -- Example, a (nearly-)sealed crucible (SiC maybe?)? O2 sensor? Gas analysis (sampling, GC, MS, etc.)? (Interestingly, ZrO2 is an oxygen ionic conductor, so might not be as attractive a crucible material in this case, as it is otherwise.) – Tim Williams Jul 21 '22 at 22:05
  • @TimWilliams I really just wanted to point out that the easiest thing to do is say that an idea won't work. In science, you are more often times right to say so, but it's those few out-of-the-box ideas that sneak through that go on to become something. If you can't get the energy up to just try, you'll never know...and like I said...it's only $30. I also have a couple 22,000uF/200V capacitors on the way for all of $30 more. I can limit the voltage with my controller at 75% which will keep it well below 200V. We'll see. – Daniel Van Antwerp Jul 21 '22 at 23:12
  • @DanielVanAntwerp Indeed; also also to say, if you're not in a lab (hobby? semi/pro?), some of those instruments might be quite hard to find indeed! So I don't blame you for looking for something cheap and simple to try, even if it has a small chance of success: it may be one of the better value options, considering. – Tim Williams Jul 21 '22 at 23:57
0

Do you plan to install the full wave bridge after the SCR phase controller? If not, the controller may not work well on pulsating DC, especially if you add capacitors. This may be obvious, but stating that you want to connect the FWB across the 240V shop supply makes me want to ask.

One thing to note is that with the SCR controller regulating at 100 VAC, it is phase firing close to peak, and thus there will be high dV/dt transients that might create RF noise which could be induced into the glass. It may also cause mechanical vibration and high pitched sound waves.

Capacitors do not work very well for phase-fired AC controllers, as they will be subjected to high frequency components that can cause heating, especially for ordinary high value capacitors designed for 120 Hz rectified power. An inductor may be much better, perhaps even in conjunction with the capacitors, but design can be tricky, and they are large, heavy, and expensive, unless you make your own from a salvaged microwave oven transformer.

If you have access to a proper neutral, could you switch over from 240 to 120 once you have reached your goal temperature? With that, the controller would be operating far from the peaks, and the high frequency components should be much reduced. I think that may have a better chance of good results - at least enough to see a difference.

PStechPaul
  • 7,195
  • 1
  • 7
  • 23