Author Topic: Pointless energy wasting North American single tap autotransformer ballasts  (Read 509 times)
WorldwideHIDCollectorUSA
Member
*****
Offline

Gender: Male
View Posts
View Gallery

HID, LPS, and preheat fluorescents forever!!!!!!


Worldwide HIDCollectorUSA
Pointless energy wasting North American single tap autotransformer ballasts « on: October 26, 2021, 02:32:07 AM » Author: WorldwideHIDCollectorUSA
Whenever I have been looking at single tap HID ballasts for the North American market, I have noticed that some single tap ballasts (usually the CWA ballasts) have a seemingly unnecessary autotransformer as part of the ballast when it is suited for a mains supply above 120v such the following 277v only 400w H33 mercury vapor CWA ballast:

https://www.ebay.com/itm/161297662850?hash=item258e170382:g:LMMAAOxyBotTaV~o


When I looked at it, it has an integral autotransformer that steps down the mains voltage from 277v to 250v to run a 400w H33 mercury vapor lamp rather than using a plain simple choke and capactor wired in series with the lamp. In my opinion, having that integral transformer on a 277v only ballast seems like an unnecessary waste of energy to me. Why is an integral autotransformer is needed for CWA circuits in this application?


Another thing that I find worse is when looking at 480v only HID ballasts, I know that a 480v mains supply is high enough to ignite almost all HID lamps on a simple choke ballast without an autotransformer in the circuit. However, I notice that some North American 480v only high pressure sodium ballasts intended for 100w S54 lamps have an integral autotransformer that steps down the mains voltage from 480v to 120v as seen with this HX ballast: https://www.ebay.com/itm/164045300476?hash=item2631dca2fc:g:iT8AAOSw9zdeKeK7

In addition, I also know that there were also some 208v only HX autotransformer ballasts designed for 175w H39 mercury vapor lamps known as the Advance 74P3313-011 postline ballast that has an integral autotransformer that steps up the mains voltage from 208v to 245v whereas a 175w H39 mercury vapor lamp can satisfactorily run on its own using a simple choke ballast on a mains voltage that is around 190v.

In contrast, I notice that some properly designed 277v preheat fluorescent ballasts are simple choke ballasts without an autotransformer that are designed to run some linear fluorescent tubes and non integrated compact fluorescent lamps that can run on a 120v mains supply with only a simple choke ballast without an autotransformer such as this Robertson 277v F20T12 preheat choke ballast:
https://www.ebay.com/itm/123627788491?hash=item1cc8ca58cb:g:kfUAAOSwV51cV2rw

Why are some of these high voltage HID ballasts such as the mentioned 480v only 100w S54 high pressure sodium ballast designed with these lossy step up and step down autotransformers whereas the mentioned 277v preheat fluorescent ballast designed for 120v OCV fluorescent lamps is a simple choke without a step down transformer?
Logged

Desire to collect various light bulbs (especially HID), control gear, and fixtures from around the world.

DISCLAIMER: THE EXPERIMENTS THAT I CONDUCT INVOLVING UNUSUAL LAMP/BALLAST COMBINATIONS SHOULD NOT BE ATTEMPTED UNLESS YOU HAVE THE PROPER KNOWLEDGE. I AM NOT RESPONSIBLE FOR ANY INJURIES.

Medved
Member
*****
Offline

Gender: Male
View Posts
View Gallery

Re: Pointless energy wasting North American single tap autotransformer ballasts « Reply #1 on: October 26, 2021, 03:57:23 AM » Author: Medved
There are many(technical) factors:

One is, many lamps not only need the specific current when the lamp operates at the rated arc voltages, but the ballast response (so the current as a function of the arc voltage) has to follow certain shape to ensure lamp thermal stability. These restriction of the shape then yield to lower OCV than the supply voltage, so some kind of transformation is necessary. Mainly important for lamps that may use saturated vapor, like HPS or many MH.

Other is the required robustness against anticipated mains voltage tolerance and fluctuation. So the use of CWA even where a series choke could be voltage wise sufficient. With a series choke, 5% of mains variation yields about 10% lamp current variation. And that is the maximum most lamps can tolerate (when maintaining the rated performance). Often the voltage at the place varies way more, so the ballast is required to suppress this variation. E.g. the CWA may keep the arc current (assume the arc voltage is constant) within 10% for the mains fluctuating as much as +/-15%, some extreme designs or more complex types like magreg are able to cover +/-30%, still without any switching.

And there is the efficiency. It is true, normally the most efficient use to be the series choke. But when the mains voltage is high but the arc one low, the apparent power the series choke has to handle could be easily larger than the sum of the apparent power all windings have to handle when first using a step down transformer and then a series choke, or mainly the HX or CWA style transformer ballast.
An example (still neglecting the load response, just taking the losses): Your mains is e.g. 480V, arc voltage 55V and arc current 2A (about the 100W HPS). A series choke would have virtually the whole 480V across it, with the 2A flowing through. That means its winding has to handle 960VA, with quality factor of 20 (pretty high for such a choke). that still means 48W losses. On the other hand when using 120V series choke ballast with its rated PFC capacitor and then a 480/120V transformer, the ballast choke has to handle 213VA, so dissipate about 10W (assuming the same Q factor). Assume the power factor is then corrected to 0.95, so the apparent power the transformer has to handle is 115VA. It has two windings, so the sum is 230VA sum of both windings. With the same Q=20 it means about 12W losses. So in total we are at 22W losses, even when using completely separated transformer to convert the 480V down to 120V, plus p4roviding insulation between the 480 and lamp circuit.
When using autotransformer for the 4870/120V conversion, the windings would have to handle just 173V, so total power losses become just barely 19W.
When using a HX autotransformer, it would have to handle 360VA in total, so power losses of about 18W. When arran ged as isolating, it would be 480VA so 24W losses. So even when using a special component for the given lamp and mains combination, it has higher losses than using standard 120V mains ballast and an universal use 480/120V transformer, both components with way wider use so most likely available at cheaper price.

And the last is not that much technical, but still very important aspect:
The more common the parts of the complete ballast are, the cheaper they are. Even when manufactured in-house of the complete ballast maker (using complete wound chokes already produced for some mainstream product to build a niche product is way cheaper than developing and manufacturing special parts for the same). So the seemingly unnecessarily complex ballast is so complex just because that way it is just assembled from parts that are readily available and not need to be special made for it, even when it seems to be less economical on e.g. material use or operating efficiency, still the mass production of its components makes that way the cheapest in the total cost over the lifetime. Plus easier to fix later on, when some part fails and need replacement.
Logged

No more selfballasted c***

Print 
© 2005-2024 Lighting-Gallery.net | SMF 2.0.19 | SMF © 2021, Simple Machines | Terms and Policies