Electronic ballast cable length outside the reccommanded specs

[R005T3r]

Hi,

I'm currently working on a project that it's purpose is to modify and update an old bromograph that has 12 fluorescent lamps in it. I've alredy reverse engegneered it, and everything was fine. However, when I've started looking for   electronic ballasts I've seen something unexpected: it appears that every manufacturer specify the cable length from the output of the ballast to the lamp should be within 1m or 2m maximum.Searching on google found little documentation about it, except from this page:
http://www.tridonic.com/com/en/faq-electronic-ballasts.asp 
that says:
Q: "How harmful is it to exceed the allowed secondary cable length?"
A: "The cable length has no influence on the light output. An overly long cable length however leads to capacitive currents too high to earth. This can cause undefined ballast shut-down and decreased heating current, which will have negative influence on the life time of the lamp. Additionally starting problems can occur and the EMC behavior is worse.
Short summary: Don't exceed the maximum allowed cable length. (the length of the wires is not really the issue, the capacitance of the wires is what counts (depends on wire type). Typical value is 1 m = 100 pF (varying from  wire type to wire type). As it is almost impossible to measure the wire capacitance we recommend to calculate with 1 m.)"

Unfortunately, I can't comply to the limits they recommend and "overly long" is quite vague. Besides, I don't want the lamps to shorten their life either. What can I do?    I have a capacitance meter should I measure the capacitance from the wire to the ground and from cable to cable? Is there another option rather than replacing the  ballasts ? Will EMC be an issue (there's a pcb with a microprocessor in it)?

[Kleinstein]

My main concern would be EMI. More cable makes it a better antenna. So with long cables it would be difficult to meat EMI standards. I would not be to concerned about reduced heating and thus change in lamp life.
It is not only the cable length / capacitance that counts. Also inductance matters - so keeping the cables close together helps reducing the inductance and also the magnetically radiates field.

[SeanB]

Just how long are the cables in the unit, that you are worried. 5m would be a pretty big exposure unit, and I guess you are going from the choke start ballasts with starters to electronic ballasts. You probably would be fine with placing the light mass ballasts closer to the lamps, probably on the base of the reflector, or just keeping the regular choke ballasts and putting in electronic starters, which have a defined preheat time then strike the lamp once.

Electronic ballasts are often used in signs, and there you might have 10m of cable to the lamps, and they solve that by simply placing it in air, not direct against metal, so you probably could get by with doing the same, or running them in PVC conduit there. If you really need long wires then using older style instant start ballasts ( which are a ferroresonant transformer with current limiting via magnetic shunts and include filament transformers as well, you need one per 2 lamps, but they have to be T12 lamps not T8 or T5 but they are not length limited much) or the choke style is the only method.

[R005T3r]

Putting the ballasts closer to the lamps is not possible because the ballasts were designed to fit inside the front panel. The worst case scenario is from the ballast to the lamp is about 2 / 2.5m.

I was opting for these electronic ballasts because they fit inside the wood panel:
http://www.tridonic.com/it/download/data_sheets/DS_PC_TC_PRO_en.pdf
mod. 2X26-42 TC PRO 70W  that are the only ones readily available at my hardware store. Lamps are T12

Actually the unit weights a ton, so I would like to use electronic ballasts also to reduce the weight and get rid of the starters too.... There's a wood plate inside the unit where there are the choke ballasts. I may use it to lift the ballasts as you suggested, but the cables could be in direct contact with some metal when they connect to the lamps, because the fittings are kept in place by a metal string.

[David Hess]

I expect that the cable length is primarily specified to meet EMI/EMC requirements.

When you route the wires between the ballast and bulbs, keep them away from metal structural elements and other wires.  Maybe run a test where you deliberately double the wire length to make sure that the ballast still function correctly under even worse conditions.

[Seekonk]

A friend has an electronic ballast on his septic system for a UV tube.  that is longer than 1 meter.

I just restored a piece of equipment and I took out the tube altogether and replaced it with LED strip lights. There aren't many applications where the LED strips don't work better.

[R005T3r]

I expect that the cable length is primarily specified to meet EMI/EMC requirements.

When you route the wires between the ballast and bulbs, keep them away from metal structural elements and other wires.  Maybe run a test where you deliberately double the wire length to make sure that the ballast still function correctly under even worse conditions.

Unfortunately, the unit is totally in metal....

[SeanB]

3M should work, just make sure the metal reflector is connected to the frame with a nice wire, and put the ground wire to all the ballasts, including the reflector as well. That should keep the RF down, and they will light all the lamps. Just wire the lamps with twisted pairs for each side of each lamp, and run the 4 wires back as close as possible to reduce the loop ( run under the reflector to the feed side), and try to put the ballasts so all lamps have roughly the same length of cable.

T12 lamps, you will struggle to get replacements, so look after them, almost nobody makes them any more unless you have a NOS box of them. As a bonus they do have a lower strike voltage and slightly lower running voltage, and a much longer life as well, and electronic ballasts do not strip cathodes as much with the gentler starting. i have some IS ballasts at work where the lamps last around a decade in service, though at the moment all the relamps are using used tubes from other fittings. Some were Westinghouse lamps, last made in 1978 in that base style.

[R005T3r]

Running cables under the reflector is impossible due to the fact that it's screwed to the frame of the unit. However, I'll try to figure something out...

My hardware store have a lot of T12 lamps, not  UV ones tought... Sylvania are still in commerce, but they are becoming quite hard to find, since anyone now is switching to LEDs (and quality UV-B leds are insanely expensive). That's also why I want to extend their life as soon as I can.

Another thing I've forgot to mention: my T12 lamps are 20W each, and there are no 20W electronic ballasts out there... The ballast my hardware store has available are the ones provided in the link above. Will the extra 6W shorten/wear the cathodes? 

[Zero999]

A friend has an electronic ballast on his septic system for a UV tube.  that is longer than 1 meter.

I just restored a piece of equipment and I took out the tube altogether and replaced it with LED strip lights. There aren't many applications where the LED strips don't work better.

Was UV tube the septic system a germicidal tube with a UVC wavelength? If so, how did you find a LED strip with a similar wavelength? UVC LEDs are normally very expensive.