Film-Tech Forum ARCHIVE: switching rectifier 101 question

This is topic switching rectifier 101 question - efficiency in forum Film Handlers' Forum at Film-Tech Forum ARCHIVE.

To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=1;t=011460

Posted by Jim Henk (Member # 3628) on 05-25-2013, 04:11 AM:

I do know that regular high reactance rectifiers (transformer, bridge diode circuit, cap, voltage regulator) are about 50% efficiency. Easier to service, but hot, heavy, and maybe even more important, they run up your electric bill something fierce.

So I'm told that switching (or switch-mode) rectifiers are a lot easier on the wallet at electric bill time, and have gotten past the first models that had high failure rates. At least that's what I've been told.

Question: Really? What kind of efficiency increase are we really talking about? I know from my reading that the rate is not even all across the timeline, but can I get some knowledge from experience here? I'm already starting to check out a Super Lumex replacement for my Christie H-10 lamphouse, (I said *starting* to look, TB. [Big Grin]

) and am wondering if I should replace my small Christie old-style rectifier (though it recently got refurbished with new wiring, all new diodes, etc) with a step forward while I'm at it.

Thanks.

Posted by Steve Guttag (Member # 268) on 05-25-2013, 05:38 AM:

HR rectifiers SHOULD be more than 50% efficient. That is, if you are driving a 1KW lamp, you should NOT be running up 2KW in electric. It is likely in the 80ish% efficient range. But that does mean...in addition to the 1KW lamp you are using on screen...you will likely feel the heat a 200-watt light bulb puts out coming off the rectifier itself.

Switch mode rectifiers are typically over 90% efficient. While at first you think...well that is only a 10% difference but in our example above of a 1KW lamp causing the rectifier to give off 200-watts of heat...with the switch mode, we've cut that in HALF.

There are a lot of factors in there, particularly in HR type rectifiers that can vary their efficiency. Including its power factor, how it is designed to handle a particular current range...etc. Look at its input current requirement or measure it and the voltage and see what power is going into your rectifier. Then measure the output voltage and current. If when you calculate V*A in does not equal V*A out, you will see what the actual efficiency is. You will not see its power factor directly there (ever notice that rectifiers and other "reactive" devices are rated in "KVA" rather than watts? VA only equals watts when the power factor equals 1. You'd have to measure the time difference to get watts but VA should be sufficient for this.

Switch mode rectifiers remain more difficult to service but they are just about everywhere in your life already...most electronic devices use them and all of the DCinema projectors use them. To give an example, a Barco DP2K-12C that can run a 2KW lamp runs on NO MORE than 15A (200-240V) for the lamp and projector combined. Compare that to a typical 2KW film system.

Posted by Gavin Lewarne (Member # 5397) on 05-25-2013, 07:44 PM:

Steve -

I put my (not terribly expensive) clamp meter around the power line going to our DP2k-12c, and at 1400w lamp power, and actively projecting, it was using a little over 10 amps (240v supply).

Not sure I totally believe that, but my meter isn't great.

Posted by Steve Guttag (Member # 268) on 05-26-2013, 12:39 AM:

It is probably pretty close. If we accept that running 2KW lamp at 200V draws no more than 15A is accurate....then we have a relationship that 3KW in equals 2KW out (remember, we are driving the extronics, fans...etc too on top of the lamp). So if the input voltage is instead 240V...then the current should drop to 12.5A. If you lower the wattage down to your 1400...we are down to a predicted 8.75A. So not counting for anything being not exactly as stated...we are within 2.5% of the reading you gave. Add in the efficiency of the rectifier, accuracy of your meter, accuracy of the stated voltages and wattage (did you measure the wattage of the lamp and in the input voltage?). I think we are in the range of believable.

Now, to put that on topic...take a typical 1.6KW HR system (most 1.6KW lamps really ran at 1.4KW)...see what that current draw was...it is going to be higher.

Posted by Monte L Fullmer (Member # 2797) on 05-26-2013, 01:07 AM:

It was known that the HR rectifers in Christie SLC consoles were horrible in drawing line current compared to a STRONG switcher of the same DC output capacity.

About anytime you send current through a transformer's primary winding at the beginning of the current string, it's going to do some major current draw, esp with the gauge of solid copper wire that makes up the construction of a transformer, which in turns, becomes a resistor.

Posted by Louis Bornwasser (Member # 3063) on 05-28-2013, 03:33 PM:

Christie rectifiers are very inefficient. It would literally pay to replace with a "normal" HR or switching rectifier in $$$ because of power.

Christie used series resistance, not series reactance to regulate. That is what the nichrome wires on the rectifier were for; note that no one else had them. Locally, power is cheap so not worth it, but New York and Boston, well worth it. Louis

Posted by Steve Guttag (Member # 268) on 05-28-2013, 04:30 PM:

Ahem....look in the bottom of a Strong 77000 series rectifier and check out those two largish coils of wire. Strong too used those "space heaters" to change a 4KW rectifier into a 3KW or a 2.5KW into a 2KW.

But really, just use a thermal probe...any heat coming out of a rectifier is power that did not turn into lighting the lamp...any heat coming off of a lamphouse or up the stack is power that didn't go to getting visible light.

Pulling the information from a Christie CC20U. It is rated to be at 15A 3-phase when powered by 208V. This translates to about 5.4KVA