I have two Advance dual-40W RS ballasts. One is a residential-only Advance Benchlite ballast, which is quite a bit smaller than the other one and doesn’t mention anything about thermal protection. The other one is a standard beefy commercial-grade Advance ballast with AdvanGuard self-resetting thermal protection. Both are magnetic ballasts, run the lamps in series, and are potted in tar (thank god it isn’t epoxy). Both of these ballasts worked beautifully at the time of removal from their fixtures although they both are a little bit noisy and get very very hot (which I think is normal because I have another one still in service that also gets very hot). I conducted the following measurements with both ballasts:

The residential-grade Advance Benchlite ballast has an OCV of almost 300V, which is what I would expect for a ballast that needs to strike across two 40W lamps.

The commercial-grade bigger Advance ballast has an OCV of only around 160V, even though it is a series ballast and has to strike across two 40W lamps in series.
This larger ballast is the one I ended up taking apart.

I drilled out the rolled edges of the bigger Advance ballast and pried the bottom off of the enclosure, and of course only saw tar. Well, very long story short (I will probably make a gallery post about this whole process), I was left at the end with the intact capacitor and transformer from inside the ballast, and I reconnected them to each other properly and was able to see everything connected together outside of the tar. I bypassed the thermal protection. Then I re-measured the output voltage as a sanity check and it was still the same, at 160V. Now at this point I could’ve tested it on 2 fluorescent tubes, but the only working ones I have were currently in fixtures and I don’t have a whole lot of luck when it comes to walking around with four-foot lengths of glass.

Now, one detail. The capacitor I mentioned from inside the ballast has two sections with one common wire. One is I think a 3.8uF, and the other is I think only 0.3uF. In reconstructing the ballast, I didn’t connect the smaller section of the capacitor because I assumed it was just for power factor across the secondary winding. But maybe this has something to do with striking? I don’t know, and if someone could explain this it would make my day because my whole workbench smells like tar right now lol.

Thank you very much!
 

The smaller capacitor puts a very low current OCV to one of the lamps to assist in starting. If you ever see how one lamp on the blue/yellow wires will have a dim glow without the other on the red/yellow, that is what this does. The older version described here (instead of 2 separate caps) is the large capacitor with three terminals...it's basically the two caps built into one.

As for the OCV, the normal measurements should range between 270v and 320v (the higher being 'cold temperature  ballast'). But, it also is effected by line voltage which ranges from 115v to 125v, typically. Anything below 260v usually indicates a problem.

With series lamps, one of them (usually the one "closer" to Neutral) is bypassed by either a resistor or a small capacitor. The idea is, when none of the lamp is ignited, the impedance of this bleeder is way lower than the impedance of the other non-ignited lamp, so practically the full voltage becomes across the "other" lamp, igniting it.
Once "the other" lamp ignites, the resistor/small capacitor impedance becomes way higher than the impedance of "the other" ignited lamp, transferring the OCV across the first lamp, igniting it as well.

But for F40 I would expect the total OCV to be in the 300V ballpark, just for operation after ignition it needs to be significantly higher the sum of arc voltages (so 220V or so).


Are you sure the secondary is not broken somewhere?
Or that you have measured the OCV between correct wires (or correct connection, given the lamps are not there)?
Also some ballasts are wired so they do not get complete power when the lamps are not completely connected (the primary connection relies on the filament conductivity; it is a form of interlock, to allow connecting lamps in a way they won't ignite the discharge when one end is still out of the socket, so a form of safety measure to protect the one who is handling the lamps), so for your measurement it maybe was not completely powered...

The smaller capacitor puts a very low current OCV to one of the lamps to assist in starting.

With series lamps, one of them (usually the one "closer" to Neutral) is bypassed by either a resistor or a small capacitor.

Thank you so much, this helps a lot. I will probably end up testing it in this way carefully with the proper lamps in place and see if they strike. I wonder why the seemingly higher quality ballast relies on this mechanism while the other cheaper one does not. Thanks again!

This mechanism needs an extra component, but ensures more reliable and better defined starting (more reliable, at the same time less too early cold starts).
But 160V for both lamps in series would be way too low to even operate, so that seems to me as impossible.

300V OCV for both in series seems to me as adequate if the "bleeder" trick is used.
Otherwise without the bleeder, I would expect something around 400V or so...

This mechanism needs an extra component,

What extra component is needed? I just tried the ballast with two working tubes, and it didn't work. If I shorted the red and blue wires briefly for the inductive pulse, the lamps struck and ran perfectly as expected, even with the low OCV. This has something to do with the series capacitor, and I looked into it a little bit here:

https://www.lighting-gallery.net/gallery/displayimage.php?album=8574&pos=1&pid=255553

I just wonder what component I missed that could have been used to strike the tubes. Or maybe this ballast is just toast.

The extra component is either a resistor or a capacitor in parallel to one lamp of the two in series. The 80nF section of the compound capacitor could well be that...
Now people may argue whether it is better to put it across the "hot side" or "cold side" lamp, it will do its basic job either way, only in some conditions better in the first and in some better in the second case...

Ahh I see, I did try that with the capacitor and just a very very dim flickery discharge appeared on the tube that wasn't bypassed with the capacitor. So I guess the ballast must have gotten messed up in the process of de-tarring it.