Recently, I’ve been trying to dig up information on the long term plans for Fullerton’s series circuits. I’m sure some of you are familiar with Fullerton, CA as they have HPS gumballs and vintage MV fixtures such as OV-12s, Unistyle 175s, M-250Rs. From what I’ve read, they have a aging lighting system that is becoming unreliable, and according to SCE the components such as the Regulated Output (RO) transformers needed to keep their system functional are no longer available.

Recently I found this official city document highlighting the difficulties the city is having with their series circuits and the potential options of upgrading their lighting system. Options include outsourcing to SCE, keeping the system city-owned but rewiring it to multiple circuits, or keeping the system city-owned but using series to multiple (STM) transformers. The city even considered reaching out to the fixture manufacturers to work with the city on the issue. Each solution has its benefits and drawbacks, so from my understanding the series circuits are here to stay for now.

Regardless of what the city does, I have already documented a number of fixtures and invite other nearby collectors to take a look at their lighting system before some decision is made. The rest of the city owned poles that aren’t on series circuits have been replaced by LEDs such as Philips Roadfocus, but I have found a EOI and some early GE ERS3s. Only time will tell what will happen to Fullerton’s ancient street lights.

Will you try to save any of the old lights?

Maybe, but I’m not sure if I can.

What surprises me, these problems didn't get adressed sooner than 60 years after the system became obsolete (when the incandescents were replaced by MV)...

Probably the estimated cost to redo the system back in the day was probably expensive, which could explain why it was differed until now.

Yeah and plus the manufacturers still made the associated gear for such a system until recently. The change from incandescent to MV and MV to HPS was fairly straightforward as you could order luminaires for those lamp types that runs on 6.6A system.

The components and wiring were probably in better condition back then so it was harder to justify the cost of a full system rewire and new luminaires compared to just buying the newer MV and HPS luminaires.

When LED came around I suppose the system aged enough that finding parts would be an issue and plus LED luminaires don't natively run on series circuits without the use of a STM transformer.

HIDs need ballasts (or current conversion transformers) on the series circuits, which brings the main reason (running directly the lower voltage, higher current, so more efficient incandescents) for the series circuit away, while the inefficiency of the regulation transformer (ut is in fact variable leakage HX teansformer, so with huge and varying reactive power on its input) would remain.
I understand why these special ballasts were made: To gradually replace the incandescents with HID at the 60 years ago time. But then I would assume the series circuits would get dismantled, together with the regular overhauls. But probably at places where the pole rust is not a problem, the overhauls are not that frequent, so the overhaul happens onoy now...

When I recently took a visit to Fullerton, I have noticed that one of their high pressure sodium gumball lanterns exhibited significant variations in brightness.

See here:

https://www.lighting-gallery.net/gallery/displayimage.php?album=6272&pos=0&pid=206411

I saw something like that on Reddit, however it was a significantly more noticeable pulsating effect.

I wonder what causes a high pressure sodium lantern to vary in brightness on a series circuit.

I wonder what causes a high pressure sodium lantern to vary in brightness on a series circuit.

Series circuit will more likely impose really constant current, what means possible thermal instability of the lamp...

Thanks for the observations. Just out of curiosity: If the city did use STM transformers, would they have to install one per circuit, or one at each pole individually?

On the design I've seen, you have to install one per pole. Usually like how the old style remote MV ballasts were installed. I suppose on a double arm pole and a lower powered luminaire, you could probably wire both luminaires to one STM transformer.

The series ballasts generally wired up in a similar way too, one per luminaire.

It always seemed interesting how prevalent series streetlight circuits were in the US. I think they were long gone here in Canada. Even the old incandescent gumballs here were multiple. 

Ahh, I see.

The STMs could be feasible for the city. I believe they might have converted one of the series poles to LED using a STM at the base of the pole as a demo.

Now the tricky part is the regulator transformers (ROs) that are between the beginning of the circuits, and the feeds from SCE’s grid. If those aren’t produced and one let’s say one fails and can’t be repaired, then the city I guess and would be forced to rewire that circuit to multiple and the STMs would be useless. If several ROs begin to show signs of failure, then I think it would make more sense to just rewire the circuits citywide to multiple, but again that’s if the ROs are a problem. I’m not really too smart in terms of how these systems work but that’s my understanding.

Yeah the regulator transformers are the main issue especially since they might be nearing the end of their service life and the same applies for the underground series wiring too.
Here's a good overview on how a basic series circuit is laid out.

The RO is the device at the beginning of the string, it basically takes the incoming power and regulates it to ensure a constant 6.6A current (CC). You can think of it being like a more precise ballast on steroids lol. Anyway the goal of series originally was to allow the use of lower voltage (incandescent) lamps which are usually more efficient than their line voltage counterparts and avoid having the lamps at the end of the run be dimmer due to voltage drop on long 120v runs. Because of the number of lamps on a series string the voltage of the system overall would be higher, typically a few kV but the voltage across each individual lamp is a lot lower, around 12-40 volts.

A series system is basically a scaled up version of those mini incandescent Christmas lights, but with a added current regulator to keep the current constant. Notice how if one bulb burns out the rest of the string will burn brighter due to the same voltage applied over a smaller number of lamps. On a series system the regulator will lower the voltage if a lamp burns out to maintain the 6.6A CC. The lamps in a string of Christmas lights have a similar film cutout as what's installed in a series streetlight to allow a failed lamp to be bypassed in the string.

When HID was introduced, they made ballasts that were designed to make use of the existing series infrastructure. There was an attempt at making special 6.6A MV lamps that operated directly off the 6.6A system but the special lamps and maintenance issues meant the isolation ballasts won out at the end. Anyway I suppose if there was a will, you could probably make a LED luminaire with a integral STM transformer but that's going to be expensive as no one really makes such a thing and a off the shelf LED luminaire probably won't have a provision to mount a additonal transformer in the housing.