I'm starting a project where i will need to light from battery in night street lighted in HPS.
So for keep the color it woul'd be stupid to use any light source and put color gels on it, because i'll loose intensity while sodium have the best lm/w.

I have success to use an LPS on 12vdc but do you think it's possible with HPS ?
Did anybody try to use HPS with an Xenon HID ballast ?

You may have trouble to find proper lamp.
Automotive MH are rated for 0.5A/75V, so you would need to find 70V arc HPS lamp...
Otherwise it would work well, these auto-ballasts regulate for constant power delivery (but with boost to ~70W for few seconds after power up; lamp should survive it without significant problem), so it should be quite thermally stable.

thanks for advices !
Ok for 70v hps, but did they exist ? It's veary hard to find voltage of decharge lamp. And automotive ballast are most of the time for 35w, so i have to find a 70v 35w bulb

Maybe stupid, but the common voltage of HPS isn't around 110-140 ? What about two ballast in serie ?
I do not want to blow a bulb

HPS arc voltage rise, as the lamp age, while at all times it should not be above 1/2 of the ballast OCV (for lag ballast) in order to have stable arc (in fact it is the arc voltage rising above that level, what cause the lamp to cycle at EOL).
In 120V areas are for higher wattage lamps (150W and above) used transformer type ballasts, where the OCV may be set arbitrarily to any value needed, so lamp standards are designed first to have optimum operating characteristic (that lead to arc voltage from ~80 to >150V) and the ballast OCV is set according to this.

For low wattages (150W and below) it is beneficial to use only series choke ballast (it could be easily half the mass and half of the losses compare to autotransformer type), but these have the mains voltage as OCV. So lamp standards are designed so, the arc voltage reach the ~1/2 OCV limit at EOL, what mean ~20% lower initially. This mean departure form the optimum lamp, but for the low wattage the difference is so small, then the energy savings on the lower ballast losses offset this.
Hence arc voltage of new 35..150W lamps for 120V market is ~55V. Note, then the 150W is sold in both variants: 55V for serial reactor as well as 80V for the transformer ballast and they are not compatible.

In the 230V market the 230V is enough as OCV, so all HPS are specified with arc voltages in the 70..100V range, to work on simple series choke ballast. This voltage is maintained even for low wattages, as ballast losses of the series reactor are mainly linked to the operating current, so it has to be kept as low as possible.

So for the 35W MH replacement you would need "european" style 35W HPS lamp, but i don't know, if such lamps were ever made.

Thanks for response Medved,

I think i understood, but i'm not so good,there is somes abreviations i don't understand.
I guess OCV is something like operating voltage and EOL is End Of Life ?

So for the 35W MH replacement you would need "european" style 35W HPS lamp, but i don't know, if such lamps were ever made.

Yes it's can be hard to find, no one constructor catalog make diference talk about designed lamp/ballast for 120v or 230v...
If i find a 35 HPS, how can i know witch voltage is design for ? If i test a lamp with ballast, what is the risk ?

I should find 150w 12v ballast

@irpyc:
OCV = Open Circuit Voltage; I is the voltage the ballast provide, when not loaded. Igniton pulses are excluded.
EOL is mostly End Of Life,
 sometimes (in context of manufacturing process and quality testing) it is used as End Of (production) Line, so "EOL test" mean test performed at each produced piece at the end of the manufacturing line, used to identify quality escapes...

35W HPS would be always marked only as "35W". It depend on the market, where it is sold: In US it would mean 55V arc, in Europe 70V arc. But when you find detailed specification, the arc voltage would be there.

The 150W HPS ballast you would have to make yourself...

It's working !

I bought a Plusrite HPS 35w, who work with S76 ballast, so i try with my xenon ballast and it's workin well. Mutch more bright than my 35 LPS en inverter, but i think it's the inverter who don't use 100% of the lamp.
I hope it will not reduce to mutch the life time of lamp...


We can use HPS as car headlight


Strange thing, when i but XLR 3 pin beetwin the ballast/lamp, with HPS it's working great, but with xenon hid it's making short circuit arc in the XLR... maybe the ballast change the intensity in fonction of lamp :/

Can we use HPS lamp without any lantern, or there is a risk of blow up ? And what about UV ?

I tried something similar once : 70 W HPS (internal start) on 240 V inductor ballast, using old computer UPS to convert 12 V to 240 V

Problems were :

First time i used too weak UPS, so wnen the glow bottle starter shorts out the UPS kicks off. It worked ok with mercury lamp though

When i used bigger UPS, it worked, but the glow bottle continued to light (probably cause of the ballast making kicks of high voltage due to being run on square wave instead of sine). The lamp looked like it lighted up OK, but then gone cycling due to the glowing starter shorting it out (probably when the lamp warms up and the arc voltage rises too high). Maybe another lamp with slightly different starter would stay lit

When i tried even bigger UPS that made sine (or closer to sine) wave everything was ok

When i used bigger UPS, it worked, but the glow bottle continued to light (probably cause of the ballast making kicks of high voltage due to being run on square wave instead of sine). The lamp looked like it lighted up OK, but then gone cycling due to the glowing starter shorting it out (probably when the lamp warms up and the arc voltage rises too high). Maybe another lamp with slightly different starter would stay lit

I saw many cases, where the ignitor continues to make kicks when the lamp already lit, in HPS fixtures with a superimposed ignitor, where the lamp loses half of it brightness and regains after several seconds, sometimes leads to cycling.
My video of one:

I seen some other stuff happening in 150 W (and possibly higher) HPS

There is the ignitor common in Israel that does not output 50 / 100 pulses per second, but a strong pulse once in 1 or few seconds (like Eltam ES 50, but i think the effect i am talking about happens with older ignitors of the same type and not with the ES 50, maybe with the "Steinitz red" ignitor common in the 1980s)

The lamp is lit at notmal brightness, but once in a second or more (same frequency as the ignitor rhythm) its brightness momentarily jumps (its hard to tell if up or down, maybe it jumps normal-up-down-normal or the like)

The lamp lights like that full-night and does not cycle. Only EOL lamps that allready are cycling get kicked off in sync with this pulse

Seen this happen in 70s / 80s gen1 Koffer 150s (running 150 W HPS, possibly upgraded from MV) and with Shaltiel "green frog" floodlights (running HPS of unknown wattage, i think 150 / 250 / 400)

Ash: i know about these ignitors that outputs only few pulses per seconds (Usually 3-5 pulses/sec).
I have two video of reignitions of cycing HPS with these type of ignitors:
Post in the forum of the first video
Post in the forum of the second video
i don't know if these type of ignitors exists outside Israel, in Asia, Europe and North American.

Yes UPS can be an idea, but yes they need a powerful UPS, but it should loss lot of power.

With xenon ballast it's working well ! But i have only 35w, there is somes 55w Xenon ballast, but i don't find any cheap one and 55w is still low.

I had somes problems on 12vdc transformer during start, with this one in can run one but not two, or with a lot of start.
With lead-acid battry is ok, but i use fuse, i need a big one.

My video of one of my 35w HPS lamp working

UPSes are quite efficient (about 80 - 90%)

Use a quality UPS with cooling fan. The cheap UPSes that dont have one are not meant to run longer than few minutes (they are running under severe overload, and designed so that the small battery they have is depleted just before something burns from overheating), so if you use cheap UPS use one with really big safety margin and upgrade the heatsink of the switching transistors / mosfets, add 12 V computer fan to it, and add 12 V computer fan for overall cooling of the UPS components

The weakest point of cheap UPS, what overheat when operated at larger battery are not the transistors, but the transformer. Transistors reach their maximum temperature in less then a minute, so they do not heat up any further. But the transformer take about a 1 hour to settle it's temperature, so with the high output power the final temperature would be way above the material could withstand. Usually the overheated transformer start to saturate it's core (it is usually designed with virtually no margin), what heavilly increase the load on transistors, leading to their destruction.

But for magnetic ballasts you would need really sinewave output UPS, as the "modified sine" (alternating positive and negative pulses with peak value equal to the nominal mains sinewave) have the gaps with zero voltage too long, so following arc reignition (with the next active pulse) is problematic. And on top of this, the "ectified average" value of this shape is ~25% less then the real sinewave, what cause ~30..50% lower then rated ballast currents (depend on ballast characteristics; assume inductive lag ballast - so series inductor, HX autotransformer and so on).
These are usually well build (as due to their complexity they belong to more expensive category anyway) and so they should survive higher battery capacity. Moreover these do not use big mains frequency transformer to convert the voltage, but a high frequency convertor, what contain only small parts and these thermally settle in few minutes, so even on the "small" battery (so with bigger one would not heat up further).

Transistors reach their maximum temperature in less then a minute, so they do not heat up any further

In many cheap UPSes the heatsink on the transistors is just a massive cube of aluminum, and does not have any fins. This makes me think that it is simply meant to absorb the heat with its thermal capacity, rather than dissipate it, and that it is meant to just slow the heating untill the battery depletes - therefore problem with larger battery

But the transformer take about a 1 hour to settle it's temperature, so with the high output power the final temperature would be way above the material could withstand

Then you can just load the transformetr less, to lower its heat output. I dont know for sure how much less, but i guess that (output VA / transformer core cross section area) is the thing to look for

If so, can i measure (VA / area) in a quality UPS meant for long work, and use this value for the wanted max. (VA / area) in the cheap UPS ? (adding a fan to blow on the transformer, like the one that exists in the good UPS)

Usually the overheated transformer start to saturate it's core (it is usually designed with virtually no margin), what heavilly increase the load on transistors, leading to their destruction.

How temperature is related to core saturation ?

But for magnetic ballasts you would need really sinewave output UPS, as the "modified sine" (alternating positive and negative pulses with peak value equal to the nominal mains sinewave) have the gaps with zero voltage too long, so following arc reignition (with the next active pulse) is problematic

I tried some combinations, and here are the results (all tests with 240V 50HZ nominal) :

Preheat 36W T8 on square (or modified with just 1 "step") wave = works ok but starter keeps glowing quite bright and is very warm after some time

Preheat 36W T8 on modified wave with 3 "steps" = works ok

Preheat PL 11W on square (or modified with just 1 "step") wave = works ok, i think the starter is glowing and heating inside

HPS 70W Internal ignitor on square (or modified with just 1 "step") wave = cycling

HPS 70W Internal ignitor on modified wave with 3 "steps" = works ok

HPS 150W external self-timing ignotor on modified wave with 3 "steps" = works ok

Bug zapper on square (or modified with just 1 "step") wave = electric net continuously arcing, lamps work ok

Bug zapper on modified wave with 3 "steps" = works ok