AC vs DC Energy Transmission for Long Distance CFLs

Thread Starter

AfdhalAtiffTan

Joined Nov 20, 2010
120
Hi guys, some questions here...

I intended to light my entire house with regular CFL, which I just found out could be operate on straight DC.

The long cable gets me thinking, if I better off rectify the AC main first, fed it to small capacitor bank, maybe 330uF, then fed it to the CFLs.

I thought I could get away with the cable inductive reactance, thus reducing power loss.

Am I on the right track?
I mean, I just read somewhere about the advantages of AC, but hey, I'm not intending to transform the voltage nor current, so DC suppose to be better.

Also, I'm thinking if this is the better way to extend the CFL's life, because of the capacitor would suppress any high spike and noise.

I'm not sure about the power factor though.
 

WBahn

Joined Mar 31, 2012
29,979
I'd be willing to wager that you will lose more in the rectification process than you will save in the line losses. Remember, whether AC or DC, you will be working with very similar voltages and currents, so the resistive losses will be very comparable. The power factor due to the cabling is going to be pretty negligible.

Watch an hour less of TV a month and you will probably save a lot more energy that this would.
 

Thread Starter

AfdhalAtiffTan

Joined Nov 20, 2010
120
Thanks for the reply.

About the power consumption, I'm intending to use this on solar bank, so every watt counts.

After some tinkering, I guess I'll go with AC, because of the potential galvanic corrosion...
 

#12

Joined Nov 30, 2010
18,224
I thought I could get away with the cable inductive reactance, thus reducing power loss.

I'm not sure about the power factor though.
The resistance of the copper makes much more difference than the inductive qualities because power line frequencies are so low.

Power Factor? On the DC side, there is no power factor because there is no phase shift. On the AC side, residential meters do not calculate a price for power factor not being equal to 1.

(This is only a partial sort of response.)
 

THE_RB

Joined Feb 11, 2008
5,438
The DC distribution would be significantly superior to run CFLs and most SMPS loads as there will be less resistive losses.

Also you made a valid point about being able to provide filtering at the source, so the loads will get much cleaner power.

It's a very novel and interesting idea, the only part that concerns me is the SAFETY aspect of making changes to your mains wiring and voltages!
 

RamaD

Joined Dec 4, 2009
328
Good CFLs do have PF correction, maybe not the cheap ones.
Resistive losses are likely to be the same irrespective of AC or DC.
With DC, the rectification process will make power factor (presented to the mains) worse!
Rectification losses will be small, due to the diode drop being low compared to the peak mains voltage (2 diode drops loss of about 2V in about 1.414 * 110 or 220V, let us say about 1%.
Significant advantage could be increase of life due to capacitor smoothing.
 

THE_RB

Joined Feb 11, 2008
5,438
...
Resistive losses are likely to be the same irrespective of AC or DC.
...
Not true. Each CFL has an internal bridge rect and cap, so the mains is converted to DC inside the CFL. Diode conduction angle is as low as 5% of the total time, so the entire power to the CFL is provided over 5% duty, this means the current supplied over the distribution wire is 20x higher than the DC running currect of the CFL inverter. That means resistive losses of the distribution wires are MUCH higher when running from the AC mains due to the fact of I squared R.

If he pre-rectifies into a large cap the distribution is of a DC voltage at 1.4x the mains voltage, this results in the CFL inverter drawing a constant and very small current, so I squared R distribution losses are much reduced.

DC power distribution is far superior to AC power distribution apart from corrosion issues.

...
With DC, the rectification process will make power factor (presented to the mains) worse!
...
That rectification is the first thing that happens inside the CFL electronics, so power factor issues are not really relevant.

The big issue is that the rectification occurs at the source, and the distribution becomes a constant small current rather than a high peak current pulse.
 

RamaD

Joined Dec 4, 2009
328
@THE_RB:
Yes, yes, I overlooked the fact that there is a rectifier inside!

The power factor controller inside the CFL electronics presents a near unity power factor load to the mains. When the CFL is powered by DC with the rectifier/capacitor, this rectifier/capacitor (followed by CFL, of course) is what is going to be presented as a load to the mains - with the rectifier conducting near the peaks with current spikes leading to a a poor power factor and rendering the PFC inside the CFL useless!
 

takao21203

Joined Apr 28, 2012
3,702
@THE_RB:
Yes, yes, I overlooked the fact that there is a rectifier inside!

The power factor controller inside the CFL electronics presents a near unity power factor load to the mains. When the CFL is powered by DC with the rectifier/capacitor, this rectifier/capacitor (followed by CFL, of course) is what is going to be presented as a load to the mains - with the rectifier conducting near the peaks with current spikes leading to a a poor power factor and rendering the PFC inside the CFL useless!
I have a bag here with various CFL PCBs- from cheap one's as well regular one's.

None of these has any means of PFC.

Maybe large CFLs have PFC (for 150W etc.)

The "large current spikes" you can easily calculate: There is a 2.2uF to 3.3 uF capacitor inside. The DC/DC converts the 400V downto about 40 to 60 volts.

And at the peak of AC mains, this capacitors gets recharged.

For a typical 20W CFL, these "large current spikes" are not much relevant.

I have run CFLs using bell wire (for instance).
 

Thread Starter

AfdhalAtiffTan

Joined Nov 20, 2010
120
@THE_RB:
Wow, your posts really enlighten me up!

@All
I guess I really need to do some measurement on this one, still waiting for my new scope to arrive.
I'll measure it with my isolation transformer of course.

I really hope that with DC, the life expectancy of the CFL would rise significantly. Those CFLs aren't cheap, even from China, at least at the place where I'm living in.
I want all my lights to be close to maintenance free, in electronic wise.

I saw countless of failing caps in cheap CFLs that I own.

I also wondering if I could make some kind of a soft-start circuit for the DC supply, maybe just a simple switched resistor ballast, filtered with inductor, but I'm afraid the CFL circuitry would die latched at low voltage.

As takao21203 said, I might measure current at each component too.
 

THE_RB

Joined Feb 11, 2008
5,438
PLEASE be very safe with whatever you do!

Regarding the life of the CFL electronics, I think you will get better life if you can run them from a reduced DC voltage. Running at 400v DC will be better for distribution losses than 250vAC, but 400vDC may not be optimal for inverter life.

The mains things that fail in the CFL are the 400v cap and the power FETs, which run at the 400v. Often these fail from heat too, so keeping the whole CFL cool would help a lot in increasing their life.

You could try running the CFL at a slightly reduced DV voltage like 320vDC or 280vDC and see if it runs the same brightness, if so that might be a good way to increase inverter life.
 
Top