0
\$\begingroup\$

enter image description here

I gained much needed help from this forum in successfully installing a full bridge rectifier on my glass kiln.

Everything has worked very well but I would like to try and filter the ripple voltage with a pair of 200V, 22000uF capacitors in parallel. I have used my process controller to limit the maximum voltage to 185V.

When I applied power to the install, the SCR voltage controller was not working. I found that a pair of superconducting fuses (63A, 240V) placed just above the SCR were blown. Picture of the system without the rectifier/capacitors is attached. They were placed below the 50A breaker (actually 60A) and above the heater.

Does the capacitor pair need to be charged up somehow prior to use in the system? It seems like they are trying to draw too much current and blowing my fuses.

\$\endgroup\$
20
  • 1
    \$\begingroup\$ Inrush current is certainly a factor, but I would be more concerned about the peak voltage (262V) that you are applying to those 200V capacitors. \$\endgroup\$
    – vir
    Commented Jul 25, 2022 at 21:06
  • \$\begingroup\$ I mentioned that I was limiting the voltage to 185V max. I used a meter to test this as DC current exiting the bridge rectifier prior to installing the capacitors. \$\endgroup\$ Commented Jul 25, 2022 at 21:13
  • 1
    \$\begingroup\$ Are you limiting the voltage to 185V AC (rms) or 185V DC? \$\endgroup\$
    – brhans
    Commented Jul 25, 2022 at 21:16
  • 1
    \$\begingroup\$ There are no capacitors in your schematic. Please draw a schematic of your exact setup. \$\endgroup\$
    – winny
    Commented Jul 25, 2022 at 21:23
  • 1
    \$\begingroup\$ "I would like to try and filter the ripple voltage with a pair of 200V, 22000uF capacitors in parallel." .. Not possible that way \$\endgroup\$ Commented Jul 25, 2022 at 21:31

2 Answers 2

3
\$\begingroup\$

Capacitor-input filtering is somewhere between not recommended, and disastrous, for thyristor based phase control. You need a very large inductor instead.

Note that Vrms or |Vavg| is not Vpk, and the capacitor charges to peak.

Easier to just use some kind of equivalent sine wave control, such as a servo controlled variac, or switching supply.

Hm, I wonder if an ESC (electronic speed controller) would do here. You would need one for a DC motor, with output filtering. Hmm, that's probably not a common combination... motor drives tend to be messy affairs (relying on the motor's own inductance for filtering). Well, it might still switch faster than mains frequency, meaning less inductance needed.

\$\endgroup\$
5
  • \$\begingroup\$ In my system, how is the SCR affected by the bridge rectifier and capacitors? It is supplying an AC current that is getting processed to "DC-like" prior to entering the heating coils. \$\endgroup\$ Commented Jul 25, 2022 at 21:34
  • 1
    \$\begingroup\$ @DanielVanAntwerp Bridge doesn't matter, beyond that it allows charging the capacitor in one direction. SCR firing into a capacitor, then, causes huge inrush current, as the capacitor was discharged by the load, then charged suddenly up to whatever voltage the SCR turns on at. With a pure resistive load, this doesn't happen and the load simply receives a variable fraction of mains. With a capacitor load, inrush current occurs, evidently enough to blow the fuse -- and indeed indicating unsafe operation. \$\endgroup\$ Commented Jul 25, 2022 at 21:59
  • \$\begingroup\$ Can I use an input current limiter? \$\endgroup\$ Commented Jul 25, 2022 at 22:03
  • \$\begingroup\$ An input current limiter such as the NTC thermistors recommended by Ametherm may help reduce the initial turn-on surge, but there will likely be repeated high current spikes when the SCR turns on near peak voltage. As I recommended in your other Q&A, why not switch over to 120 VAC line voltage, where the controller will switch on near the zero crossing. There will also be a lot less dI/dt and dV/dt noise, which also may cause mechanical vibration and induced RF that could be the cause of your problems discussed elsewhere. \$\endgroup\$
    – PStechPaul
    Commented Jul 25, 2022 at 23:05
  • \$\begingroup\$ @PStechPaul Thanks for your input. My kiln controller, like most any glass kiln, is built around 240 volts. The (expensive) SCR (DC10-24P0-0000 Watlow) needs 240V input. If 120V did not work I would not want to switch over just to find out. The system is fine while holding 2150-2250F at about 100-125V but sometimes it asks for more power when it thinks it needs it. I don't want to limit the power if a 240V system can be configured. I've described my setup to Ametherm and due to the high current he also mentioned adding a relay bypass. I will ask about the current spikes at peak voltage. \$\endgroup\$ Commented Jul 26, 2022 at 0:07
1
\$\begingroup\$

I have made three simulations for the OP's circuit, assuming a resistive load of 20 ohms (2880 watts at 240V) and a phase angle of 90 degrees. The first one shows what we have without any filtering:

Unfiltered DC output

Notice that the output is 167 VRMS and 106 VDC (Average). This produces 1.41 kW for the heating element.

The next simulation shows what happens when you put 44,000 uF of capacitance on the output of the bridge rectifier.

44,000 uF capacitor on output

Note that the first current pulse is almost 6000 amps, and then as the capacitors charge up to peak value of 321 VDC the pulses are about 100A. This is a dangerous overvoltage on 200V capacitors, and it produces 5 kW to the heating element.

The third simulation adds a 10 mH choke, which is probably about what you could get from a microwave oven transformer.

10 mH inductor and 44,000 uF capacitor on output

Here the maximum current surge is 230A and the output stabilizes at about 195 VDC which provides about 1.9 kW to the heater.

\$\endgroup\$
20
  • \$\begingroup\$ Well done. I agree that a choke is the answer. Does your C1 model have any ESR? What is the steady state voltage ripple? \$\endgroup\$
    – Mattman944
    Commented Jul 26, 2022 at 7:17
  • \$\begingroup\$ No ESR, so realistic value might be 200 mOhms and thus 1700A. Ripple is about 1.14V P-P. I hope the OP takes note of this and adds a choke. And also he should connect the capacitors in series with 10k 5W bleeder/balancing resistors to get 400 VDC rating. \$\endgroup\$
    – PStechPaul
    Commented Jul 26, 2022 at 7:28
  • \$\begingroup\$ WOW! Thanks for putting so much time into this! I am in the process of specing a 50A current limiter placed above the bridge \$\endgroup\$ Commented Jul 26, 2022 at 14:44
  • \$\begingroup\$ The actual element setup is 22 AMP-2600 WATT (EA) 4 IN KILN, 2 SETS OF 2 IN SERIES. How would these current surges ever be realized with a limiter in place above the bridge. The elements are at a fixed ~50A and if I limit the current ask of the capacitors...problem solved? \$\endgroup\$ Commented Jul 26, 2022 at 14:47
  • \$\begingroup\$ Would these resistors work? amazon.com/10pcs-Royal-Ohm-Sand-Resistor/dp/B00VVZCPBU Do they connect in series between the capacitors? \$\endgroup\$ Commented Jul 26, 2022 at 15:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.