Small UK owned streetlight design company cannot design own LED drivers?

[james_s]

I have some low pressure sodium lamps, I suppose I could light one up and put it near a photocell connected to an oscilloscope to find out if the output drops to zero.

Incidentally, I found some years ago that instant-start electronic T8 fluorescent ballasts work very well to power LPS (SOX) lamps, I used a 35W one for several years on a Philips twin T8 ballast with both outputs in parallel.

[Zucca]

Thanks treez, it is always a pleasure to read your posts.

[Kjelt]

Compete with the major electronics corporations
The other alternative is to actually design an offline, isolated LED streetlight driver with electrolytic capacitors and try and compete with the major electronics corporations such as Philips, Tridonic, Osram, Samsung etc etc…………….
This is surely hopeless…    ..there’s no way a small company would be able to get  the kind of component cost-downs that the big electronics corporations can. Not only that, but  the mainland European  electronics corporations, have benefit-clique style relations with the European semiconductor fabrication plants, so they can get components far cheaper than a UK company would be able to. (there are no major semiconductor fabrication plants in the UK, neither are there any  significant  UK owned electronics component makers).

We have talked about this before and your argument is false IMO.
The problem is not the cost of the components because that would be true with drivers costing <$10-20
The drivers you mention have a huge profitmargin for that company, BOM would be $15 but selling price triple even more when low numbers.
So you have margin enough to design and build something.

Where you will face problems is the following:

- software:
The software stack costed that company many manyears to develop and improve.
Only a validated compliant DALI stack alone would be a huge undertaking as you have found out in the meanwhile.
There are many so called SW engineers that have a 5 minute look at the spec and say oh easy.
Talk to someone who actually has worked on it , and he will say: Oh my goooood DALI it is a snakepit of exceptions, holes in the standard and you need a testing facility to see if all masters will work with your slave, don't start on it unless you want to go crazy.

Look at all the SW tunable options in these drivers, they can calculate when the sun goes up them selves just from the automated power on/off cycle from certain countries, those countries WANT that it is not an option it is an requirement. You can programm them to dim to certain levels automatically etc. etc.
Look at the accompanying PC software , there are 3 programmers working 5 days a week on that alone.

- hardware:
Open one of those drivers up and see what is in it, about 200 components, believe me none is too much or they would have taken it out earlier with other costdown programms. The big manufacturers are also always looking to lower the price in order to better compete with the chinese.

- thermal design:
Elco's you need, so you need to design a way to keep them relative cool without melting the rest of the electronics, perhaps 20 pcb layouts to test the optimal one? Or copy it from the big manufacturers, the chinese do it so can you.

- weather proof: 
Lightning protection, moist protection, go ahead build something that lasts 20 years, it is not that easy.

Then margins and profit, look at the profit warnings from Signify (was Ph. Lighting) and their sales and you see the margins are evaporating although they have a lot of bureaucratic fat in their company burning profits.
Many companies buy chinese drivers , they don't guarantee their product for 10 years anymore. The customer is the B-tester and the sucker to buy new stuff in 6 years or so, no-one cares anymore.
It is tax money who cares, that is how people are thinking today and you need to adapt as a company or go belly-up.

So that company buying those drivers probably with large discounts at large numbers is actually pretty smart unless they have an electronics+software+thermal+mech. design department summing of 20 or more persons currently eating out of their noses but I guess they don't.
Do the math 20 persons 2 years and then how many drivers do you need to sell to break even ?

[tooki]

I think flicker is potentially very significant with moving objects like vehicles. Has anyone seen cars with LED tail lights that flicker like crazy when the car is moving? It drives you nuts. Imagine if that happened to every car driving under street lights?

I thought I was the only one who got wound up by flickering tail lights!

Oh you DEFINITELY are not! I’ve ranted about this on another thread here a while ago...

[CJay]

I think flicker is potentially very significant with moving objects like vehicles. Has anyone seen cars with LED tail lights that flicker like crazy when the car is moving? It drives you nuts. Imagine if that happened to every car driving under street lights?

I thought I was the only one who got wound up by flickering tail lights!

Oh you DEFINITELY are not! I’ve ranted about this on another thread here a while ago...

Flickering tail lights, brake lights, direction indicators and those bloody VAG chasing indicators drive me nuts.

[Kjelt]

Flickering or not, automatic gear cars waiting for a traffic light breaking continuously so the break lights blind you is a major PITA. They should dim after a few seconds IMO, when there is a need to signal upcoming traffic a small action would re-establish the full lights.

[Nusa]

...the public have been whipped up into such a furore of fear of high levels of 100Hz flicker....

I bet if you stopped 10 members of the public on a busy street and asked them, 9 of them would have no clue what you're talking about. And the 10th probably would not be afraid of it.

[ocset]

    ..the public have been whipped up into such a furore of fear of high levels of 100Hz flicker....
I bet if you stopped 10 members of the public on a busy street and asked them, 9 of them would have no clue what you're talking about. And the 10th probably would not be afraid of it.

Thanks, by "public", i should have said, i mean the ones who are in charge of buying public lighting...eg streetlighting....they think 100hz flicker is a killer...because they have read European Union regulations  (EN 12464) which literally ban 100Hz flicker where the light goes to zero every 10ms…at least its banned for indoor use….due to the fact that it causes “headaches , tiredness and nausea”….i don’t agree, but it makes no odds what I think.
The fact that its banned in this way means streetlight buyers simply wont touch it…even though they are outdoor lighting….its a total knee jerk reaction ,and totally without logical basis , but that’s the way things are.

There are plenty of installed streetlights with 100Hz flicker going to zero every 10ms, and nobody ever complains about them...but this is the problem...once something is written into an EU regulation, the world stops in mindless abeyance of it.

[In Vacuo Veritas]

Just admit you're trying to design a Close Encounters lightboard to get the Mother Ship to come pick you up.

[chris_leyson]

BMW 5 series, for example, uses PWM do drive tail lights and brake lights to keep the apparent light level constant. Can't remember what the PWM frequency is but I was quite surprised to find it was quite low.

[T3sl4co1l]

Because, hypothetically, it could act like a capacitor, or charge storage device, and the electric field in the tube may take time to discharge?

Because, hypothetically, the electrons surrounding the sodium atoms that have been raised to a higher potential by the electric field may take time to decay back to the ground state (there may be a time constant for that decay).

I do not know the physics, and the above hypotheses may be wrong, but I do not see a basis for simply ignoring the possibility without evidence to the contrary.

To indulge:

1. Nah, no charge, it's a heated ceramic tube with ionized stuff inside, so the resistance will stay fairly low during a cycle (compared to cold resistance that is).  So, microseconds there -- limited more by the driver transformer inductance or capacitance, say.

2. Deionization time is definitely a thing.  It's not a charge separation thing (the plasma is very nearly neutral at all times and in all places, except for thin boundary regions near the electrodes, if applicable).  It's more of a thermal mass thing.  Plasma is the "4th state of matter", though it doesn't have a sharp phase change as the other states do (it's not a first order transition, maybe not even a 2nd order transition, I'm not sure).  Basically, it takes time for it to cool down, that is, for the ionized atoms to recombine and give off whatever energy they do (which for sodium lamps, happens to be a very large fraction of emitted 589nm light!).

There are various factors for DI time, and I don't know much about it.  Heavy ions and low densities tend to take longer.  Example, low pressure xenon and mercury thyratrons are typically ~1ms.  (Presumably, you can see the glow discharge dissipate over the same time scale.  Hmm, I should give that a try.)  Neon lamps, more like 100us.  Hydrogen thyratrons, us to ns.

Another way you can experience recombination time: sodium in a flame.  A flame is hydrocarbon radicals and various intermediate combustion products, all emitting at various wavelengths (with the CH. and related species being the most notable, in the blue-green range, for low C:H ratio (gas fuel), neutral flames.  Well, clearly those reactions are delivering a good bit of energy, if they're sometimes giving off blue light!  Introduce some sodium atoms and they'll pick up that energy no problem.  An ionic flame is also conductive, and, yes, it can even be used as a triode!  Well, both the conductivity and light emission only last as long as the atoms have enough energy to do so, so you see a streak of bright orange glow in the flame, until it cools (or finishes reacting) enough to give essentially no more emission.  You can calculate the flame velocity and see how long the sodium is (partially) ionized.   Though this will probably give an upper limit (a relatively large value like 10s of ms), because all the other things are reacting much slower than the sodium atoms themselves.

So, that's that.

One could settle this very easily and certainly by watching the spectrum of the lamp under these conditions -- dollars to donuts, it goes from sodium d-line to black body (orange hot) when switched off.   Another point in that favor: the afterglow looks damn like a thermal time constant (seconds, while nothing could possibly be storing that much electrical energy).

Oh, hey, so, on a related subject, long-life phosphors are kind of like very slow recombination.  Similar physics (well, hand-wavingly so), but in solid state instead, of course.   There are actually a few gasses with very long transition times, like singlet oxygen (which emits at deep red / near IR), around 30 seconds half life when prepared pure, IIRC.  This is possible due to spin transition rules prohibiting the singlet-triplet transition in molecular oxygen (triplet ground state), the lack of impurities with which to exchange those spins, and the slow reaction rate on container walls (because surface area is small compared to the volume).  None of these are relevant to the contents of a sodium bulb, but if you don't know that, you're actually not at all wrong to suspect that it might be possible, because, in fact, it is; just that it's limited to much more contrived conditions and specific molecules.

Tim

[T3sl4co1l]

One way to make flicker less perceptible is to increase the frequency into the kHz range. That also means smaller capacitors, and as such you might not need electrolytics.

Heh, about that--

Back when this was a hot story, https://en.wikipedia.org/wiki/Little_Box_Challenge I chatted with a guy who was thinking about solving the problem with some combination of multi-level inverters.  Well, that's fine for the AC output, but you still have to store the DC link energy somewhere, to meet the input ripple requirement (which, innocent as it seemed, on closer inspection, turned out to be about half the challenge of the design!).  He brushed that off, which worried me...  You simply can't violate the conservation of energy.

So, you can PWM at higher frequencies, sure, but now it's pulsing at ~kHz, and you will find it's "tone bursting" at 100Hz.  It's uh, not really helped all that much.

You can at least make a few optimizations, give or take how much complexity you want to deal with.

The winning design used a capacitor multiplier of sorts: a GaN bridge connected to an LC network, effectively the voltage on C is multiplied (via PWM switching) to make it look bigger.  Energy is conserved, so for, say, 50V of input ripple, 400V of ripple appears on the cap.  They also used a very boutique capacitor, an electret with built-in bias voltage, such that its C(V) curve peaks around DC link voltage (~400V) (whereas any conventional 1812 X7R chip cap you'll buy, is about 20% of nominal rating or less at that voltage!).  The conventional alternative would've been an electrolytic capacitor literally the full size of their finished box!

FWIW, C0G ceramics (if you can handle the price) and films (if you can handle the space) are some of the best on energy density, at high voltages (say, >> 200V).  They compare favorably to electrolytics, but, they can be harder to use due to the much smaller values available, and the need for higher voltages.

Tim

[Gyro]

BMW 5 series, for example, uses PWM do drive tail lights and brake lights to keep the apparent light level constant. Can't remember what the PWM frequency is but I was quite surprised to find it was quite low.

They had clever features in the past too. On mine, back in filament days, they regulated the lamp voltage (presumably PWM) and if a bulb failed the system would try to make up for the loss by using the next bulb to mimic it, eg. dimmed stop light for rear tail.

No flicker in those days!

[james_s]

Thanks, by "public", i should have said, i mean the ones who are in charge of buying public lighting...eg streetlighting....they think 100hz flicker is a killer...because they have read European Union regulations  (EN 12464) which literally ban 100Hz flicker where the light goes to zero every 10ms…at least its banned for indoor use….due to the fact that it causes “headaches , tiredness and nausea”….i don’t agree, but it makes no odds what I think.
The fact that its banned in this way means streetlight buyers simply wont touch it…even though they are outdoor lighting….its a total knee jerk reaction ,and totally without logical basis , but that’s the way things are.

There are plenty of installed streetlights with 100Hz flicker going to zero every 10ms, and nobody ever complains about them...but this is the problem...once something is written into an EU regulation, the world stops in mindless abeyance of it.

Dude, we're here right now complaining about them! Get your head out of the sand, it's not a matter of fear, it's an absolute fact that some of us perceive the flicker and find it bothersome. What part of this do you not understand? Are you just hoping that if you say enough times that nobody is bothered by it that it will magically become true? You aren't bothered by the flicker, why are you so convinced that it isn't bothersome to anyone else? How many people have you surveyed to make this determination? It's not some invisible impossible to prove thing, put a bunch of light sources in front of me and I'll tell you which ones are flickering at 100Hz and which ones are not, I'd put a significant wager on getting it 100% correct.

[Kjelt]

Just to make sure do you know for reality it is the 100Hz flicker you have problems with?
Al the TL (tube lights) and Sodium pressure lamps on conventional ballasts were 100Hz (120Hz in US) and no-one complained about them till they were EOL.
At that point the cathodes of the Sodium lamps and the filaments of the TLs are worn and you get sub-100Hz flicker which indeed is very annoying.

Modern electronic drivers use around 400Hz frequency AFAIK.

[tggzzz]

Treez starts threads in which his company is having problems with the reliability of SMPS in a remote (Balkan?) outdoor installation, "his friend" is condemned to not follow his career of choice (building SMPSs), and 100Hz anti-flicker specifications are a "scare tactic" for "no sane reason".

Common themes are that measurements are suggested but not made, and that outside forces (technical, commercial, political) are preventing him and "his friend" from going down their chosen path.

I think treez should look inward, and would have benefitted from being taught by Prof Eric Laithwite at Imperial College. Why? On principle in the final year exams he set for undergraduates, he would include three types of questions:

• one question could be answered by everyone that had attended and understood his lectures - sufficient for a pass degree
• some could be answered by people that had explored beyond his course material - sufficient for a first class honours degree
• one couldn't be answered in the time available - and should be avoided at all costs

and he expected that his undergraduates could determine which was which and avoid the impossible question. If they couldn't, then they deserved not to get good marks because they probably wouldn't be successful engineers.

[Kjelt]

There seems to be a common thread here......................................

In his defense this is probably his first or one of his first jobs and he ended at a strange company claiming to be A but acting like B.
Many small companies have their quirks and he is probably better off with a somewhat larger firm that have enough people to get help and support from esp. when the bosses are a bit clueless on their technology capabilities and what it takes to build something that seems so simple.
Although outdoor lighting seems relatively simple, it is not, there are many standards to adhere to and they often differ per country.
Then there are these horrible conditions of the offered mains that fluctuate and induced voltages on the control lines and you can go on and on.
Big companies can also get burned by these things and have to replace 1000 fixtures in some hoolaboola country they underestimated. For those big companies that is a tax write off they easily can survive, small companies can go belly up on bad projects like that.

[tggzzz]

There seems to be a common thread here......................................

In his defense this is probably his first or one of his first jobs and he ended at a strange company claiming to be A but acting like B.
Many small companies have their quirks and he is probably better off with a somewhat larger firm that have enough people to get help and support from esp. when the bosses are a bit clueless on their technology capabilities and what it takes to build something that seems so simple.
Although outdoor lighting seems relatively simple, it is not, there are many standards to adhere to and they often differ per country.
Then there are these horrible conditions of the offered mains that fluctuate and induced voltages on the control lines and you can go on and on.
Big companies can also get burned by these things and have to replace 1000 fixtures in some hoolaboola country they underestimated. For those big companies that is a tax write off they easily can survive, small companies can go belly up on bad projects like that.

All understood and accepted. In particular I'm sure the points about mains and regulations are accurate and relevant.

My statements, which are obviously very much a precis and can therefore be inaccurate in various ways (especially "by omission") are based only on what I have read here. I have no other knowledge of treez' background, situation, and future options. Consequently, while I do have some sympathy with him, I can't offer concrete suggestions for his future - only he can do that.

Having said that, I do think his approach to problem solving (as visible on this forum), could be improved. I also think he needs to find a way to take a "step back" and think about where he wants to be in the long term. That is often difficult when faced with short term pressures - but everybody must do that occasionally.

Repeatedly trying the same thing and expecting a different outcome is something softies tend to do more than electronic engineers

[Kjelt]

Agreed, I also had some questionmarks about him asking professional work related questions on a public forum.
Sometimes it might be ok'ish esp. when you don't have colleagues with the same line of work and you like a 2nd opinion.
Since the information is retained "for ever" it could become an issue in the future.  When I was a student in 1996 I asked a question online in a electronics newsgroup, you can still find it under my name, that is quite weird.

[bob225]

Flickering or not, automatic gear cars waiting for a traffic light breaking continuously so the break lights blind you is a major PITA. They should dim after a few seconds IMO, when there is a need to signal upcoming traffic a small action would re-establish the full lights.

ot. Imho that is down to bad/lazy driving practices, If your stopped for more 20-30 seconds drop it on Neutral or park

[CJay]

Flickering or not, automatic gear cars waiting for a traffic light breaking continuously so the break lights blind you is a major PITA. They should dim after a few seconds IMO, when there is a need to signal upcoming traffic a small action would re-establish the full lights.

ot. Imho that is down to bad/lazy driving practices, If your stopped for more 20-30 seconds drop it on Neutral or park

My partner's car helps create those lazy practices, it automatically puts the handbrake on when you stop but it does mean the ridiculously bright brake lights don't dazzle people

[Kalvin]

It is still surprising me after all these years that Treez has yet to revere-engineer any of the competitors' products and take a look at how the things have been done in products that are sold to customers and actually work.

[Jeroen3]

Look at all the SW tunable options in these drivers, they can calculate when the sun goes up them selves just from the automated power on/off cycle from certain countries, those countries WANT that it is not an option it is an requirement. You can programm them to dim to certain levels automatically etc. etc.
Look at the accompanying PC software , there are 3 programmers working 5 days a week on that alone.

Interesting... Do you have more info on this?

ot. Imho that is down to bad/lazy driving practices, If your stopped for more 20-30 seconds drop it on Neutral or park

That cars creep forward when releasing the brake is also deprecated. Start by fixing that.

[T3sl4co1l]

There seems to be a common thread here......................................

In his defense this is probably his first or one of his first jobs and he ended at a strange company claiming to be A but acting like B.

I think he's been in more or less the same field for about a decade now, changing jobs at an average turnover rate (a couple years?).  It could very well be that he's only found companies that meet that description, or that he's been at the same one all along and neither management nor their employee(s) (or at least just the one) realizes what they want to do, and how.

We owe this curiously long experience to OP's affinity for online forums, at least three over the years I'm aware of.  Always seems to keep popping up.

Tim

[Kjelt]

Look at all the SW tunable options in these drivers, they can calculate when the sun goes up them selves just from the automated power on/off cycle from certain countries, those countries WANT that it is not an option it is an requirement. You can programm them to dim to certain levels automatically etc. etc.
Look at the accompanying PC software , there are 3 programmers working 5 days a week on that alone.

Interesting... Do you have more info on this? 

What exactly ?
The configuration software differs per manufacturer, Philips now called Signify has the MultiOne tool:
http://www.lighting.philips.co.uk/oem-emea/products/philips-multione-configurator
For the dimming it is software, depends if the driver still has power like with Dali it can then calculate by the on off commands how long it is dark and find the right time spot to dim (as configured). With hard on off,  normally the streetlights are turned on automatically from a central control when it becomes dark, fixed programmed for the year and turned off when it becomes daylight. So software knows how many hours it is on , stores this information for a couple of cycles in NV memory and from a few cycles it can calculate what period in the year it is at and start to dim according to the programmed configuration. It works something like that.