About to take delivery

New to the forum? Feel free to introduce yourself / your cars. Also have a look for any frequently asked questions and information.
User avatar
MTXM
Citroen God!
Posts: 4895
Joined: Mon Apr 20, 2009 8:45 pm
Location: Aylsham Norfolk

Re: About to take delivery

Post by MTXM » Wed Dec 27, 2017 7:10 pm

Oh yes Gabor you are probably right and my apologies to Dean! Regards, Matthew T.
1989 V6 Exclusive (Poland car) - Now living in a local Motor Museum!
1990 V6sei auto (grey auto)
1990 V6sei manual (gold car)
1990 V6.24 Pallas (Germany car)
1990 V6.24v (Scotland car)
Other previous XM sold and broken too many to mention!

User avatar
White Exec
Citroen God!
Posts: 6642
Joined: Fri Mar 11, 2011 7:38 pm
Orga / RP numbers: RP7165
1996 2.5TD saloon, Exclusive, Polar White
1992 BX19D Millesime, Sable Phenicien
Location: ex-Ealing, Cheshire, W.Sussex & Surrey. Now living in Sayalonga (Malaga, Spain)

Re: About to take delivery

Post by White Exec » Wed Dec 27, 2017 7:46 pm

The most efficient shape of reflector for projecting a point source of light forward (as a parallel beam) is a parabola, circular in section when viewed from the front, with lamp filament at the geometric focus of the reflector. As such, it would constitute a perfect 'spotlight', with the beam diameter approximately the same as the circular diameter of the reflector's front edge.

In practice, some divergence (spreading) of the beam usually happens, and is desirable to provide useful "coverage", rather than a narrow 'pencil beam'. Typical examples would be anti-aircraft searchlights and theatrical spotlights.

Instead of the beam being circular (i.e. casting a circle of light), it can be "profiled" (shaped) by inserting a metal plate in front of the lamp, with a hole cut in it corresponding to the beam shape required, and with a lens in front of that, which is focussed on the metal plate cut-out. These metal plates can be known as gobos, iris-diaphragms, or baffles, depending on their application. In the case of a 35mm slide projector, the slide/transparency frame itself provides the beam shape (the hard edge of the projected picture), with the projector's lens focusing it and the transparency on to the screen.

As soon as a beam shaper or baffle is inserted, a certain amount of the available light output tends to get blocked off, unless some very sophisticated optics are incorporated to prevent it.

In the case of traditional car headlights, the requirement for main beam is fairly simple (something close to a concentrated and gently diverging forward beam, with a bit of edge scatter to illuminate the periphery), and not difficult to achieve efficiently. XM main beam is a fair example. Forward 'dazzle' is not an issue, because of the circumstances in which main beam is expected to be used.

For dip beam, the requirement is for a flat-top beam, with the necessary cut-off to prevent dazzling on-comers. Dip also needs to illuminate the road surface in front of the car (the 'apron'), with an even wash of light from left to right, extending as far as the horizontal cut-off. This roughly equates to a main beam, with additional divergence, with the top half of the spreading beam removed: i.e. a roughly semi-circular beam shape. This deletion of the top half of the beam is typically achieved by preventing the lamp from throwing light on to the bottom half of the main reflector (where it would tend to throw light upwards), and allowing light to bounce off the top half of the reflector, which puts it on to the road surface. In the case of an H1 lamp, the blanking baffle is external to the bulb; in the case of an H4 dual-filament lamp, the dip beam filament is both shifted forwards, and has a small metal cup below it.

As Matt points out, there is a metal baffle plate in front of the bulb itself. This is a common feature of both H1-type headlights and foglights, and its job is simply to prevent light from the filament going uncontrollably forwards in all directions, which would create dazzle in an otherwise controlled and cut-off dipped beam. Painting it black doesn't help efficiency!

Back with overall headlight shape, circular is best (with parabolic reflector), and these date back to the very first automobiles. Lucas' famous 7" and 5.75" circular headlights were so-called 'improved' by becoming all glass 'sealed beam' units, until Hella's 7" and 5.75" metal reflector and glass fronted units blew them out of the water. All of these units were circular and essentially parabolic.

Then the fun started with body styling. From the point of view of optical efficiency, square is ok if the parabolic reflector is retained behind it (eg early Discovery). The same goes for large rectangular (eg BX). Circular can be 'squashed' towards elliptical (eg Jaguar XJS), provided, again, provided it's not overdone.

In the case of XM (and Xantia, and iirc Xanae and a few other design exercises) the result was a fairly narrow slit, of restricted height, containing essentially small and flattened reflector units. XM main beam boasts a full parabolic reflector (of two curvatures) and so works well enough. But the dipped beam unit is handicapped not just by its shape, but by the absence (as pointed out above) of sections of reflector (above and below the lamp), a non-continuous reflector surface, and the expected cut-off and forward baffling. It may have been 'computer designed' but it was obviously not successful.

I'm not convinced by the suggestion that Mk.2 LHD headlights on dip were significantly better. I have a pair of such here (bought when we brought the car to Spain), and they were fitted. The dip beam pattern (if you could call it that) was just dreadful: two very bright spots of light just in front of the car, and gloom to all sides, even with the horizontal cut off correctly set. No even spread across the tarmac just in front of the car, or of the kerb-sides. In the end, I opted for the car's OE Mk.2 units, with non-yellowed fresnels flipped into LHD position.
Chris
1996 XM 2.5TD Exclusive RP7165 Polar White
1992 BX19D Millesime RP5800 Sable
1989 BX19RD Delage Red Deceased; 1998 ZX 1.9D Avantage auto Triton Green Company car 1998..2001; 2001 Xantia 1.8i auto Wicked Red Company car 2001..2003

User avatar
djg
Could do a 2.1 headgasket
Posts: 1181
Joined: Sun Jan 19, 2014 9:28 am
Orga / RP numbers: 7261
Location: Budapest, Hungary
Contact:

Re: About to take delivery

Post by djg » Wed Dec 27, 2017 7:59 pm

Ray's projector Xenon upgrade is really the best way out for RHD users, I think. Your regulation is saner than the one over here, so the modification is legal.
Bye,
Gábor

1996 XM 2.5 Turbo D VSX (LHD, Bleu Mauritius)
1986 BX 19 TRI Break (LHD, Rouge Vallelunga) deceased
1996 AX 14 D (LHD, Alaska White) in family

User avatar
White Exec
Citroen God!
Posts: 6642
Joined: Fri Mar 11, 2011 7:38 pm
Orga / RP numbers: RP7165
1996 2.5TD saloon, Exclusive, Polar White
1992 BX19D Millesime, Sable Phenicien
Location: ex-Ealing, Cheshire, W.Sussex & Surrey. Now living in Sayalonga (Malaga, Spain)

Re: About to take delivery

Post by White Exec » Wed Dec 27, 2017 8:37 pm

Should have a chance at seeing Ray's mod before long!
Chris
1996 XM 2.5TD Exclusive RP7165 Polar White
1992 BX19D Millesime RP5800 Sable
1989 BX19RD Delage Red Deceased; 1998 ZX 1.9D Avantage auto Triton Green Company car 1998..2001; 2001 Xantia 1.8i auto Wicked Red Company car 2001..2003

jonhopper
Can find the S1 radio
Posts: 26
Joined: Wed Oct 18, 2017 1:52 pm

Re: About to take delivery

Post by jonhopper » Wed Dec 27, 2017 9:33 pm

White Exec wrote:The most efficient shape of reflector for projecting a point source of light forward (as a parallel beam) is a parabola, circular in section when viewed from the front, with lamp filament at the geometric focus of the reflector. As such, it would constitute a perfect 'spotlight', with the beam diameter approximately the same as the circular diameter of the reflector's front edge.

In practice, some divergence (spreading) of the beam usually happens, and is desirable to provide useful "coverage", rather than a narrow 'pencil beam'. Typical examples would be anti-aircraft searchlights and theatrical spotlights.

Instead of the beam being circular (i.e. casting a circle of light), it can be "profiled" (shaped) by inserting a metal plate in front of the lamp, with a hole cut in it corresponding to the beam shape required, and with a lens in front of that, which is focussed on the metal plate cut-out. These metal plates can be known as gobos, iris-diaphragms, or baffles, depending on their application. In the case of a 35mm slide projector, the slide/transparency frame itself provides the beam shape (the hard edge of the projected picture), with the projector's lens focusing it and the transparency on to the screen.

As soon as a beam shaper or baffle is inserted, a certain amount of the available light output tends to get blocked off, unless some very sophisticated optics are incorporated to prevent it.

In the case of traditional car headlights, the requirement for main beam is fairly simple (something close to a concentrated and gently diverging forward beam, with a bit of edge scatter to illuminate the periphery), and not difficult to achieve efficiently. XM main beam is a fair example. Forward 'dazzle' is not an issue, because of the circumstances in which main beam is expected to be used.

For dip beam, the requirement is for a flat-top beam, with the necessary cut-off to prevent dazzling on-comers. Dip also needs to illuminate the road surface in front of the car (the 'apron'), with an even wash of light from left to right, extending as far as the horizontal cut-off. This roughly equates to a main beam, with additional divergence, with the top half of the spreading beam removed: i.e. a roughly semi-circular beam shape. This deletion of the top half of the beam is typically achieved by preventing the lamp from throwing light on to the bottom half of the main reflector (where it would tend to throw light upwards), and allowing light to bounce off the top half of the reflector, which puts it on to the road surface. In the case of an H1 lamp, the blanking baffle is external to the bulb; in the case of an H4 dual-filament lamp, the dip beam filament is both shifted forwards, and has a small metal cup below it.

As Matt points out, there is a metal baffle plate in front of the bulb itself. This is a common feature of both H1-type headlights and foglights, and its job is simply to prevent light from the filament going uncontrollably forwards in all directions, which would create dazzle in an otherwise controlled and cut-off dipped beam. Painting it black doesn't help efficiency!

Back with overall headlight shape, circular is best (with parabolic reflector), and these date back to the very first automobiles. Lucas' famous 7" and 5.75" circular headlights were so-called 'improved' by becoming all glass 'sealed beam' units, until Hella's 7" and 5.75" metal reflector and glass fronted units blew them out of the water. All of these units were circular and essentially parabolic.

Then the fun started with body styling. From the point of view of optical efficiency, square is ok if the parabolic reflector is retained behind it (eg early Discovery). The same goes for large rectangular (eg BX). Circular can be 'squashed' towards elliptical (eg Jaguar XJS), provided, again, provided it's not overdone.

In the case of XM (and Xantia, and iirc Xanae and a few other design exercises) the result was a fairly narrow slit, of restricted height, containing essentially small and flattened reflector units. XM main beam boasts a full parabolic reflector (of two curvatures) and so works well enough. But the dipped beam unit is handicapped not just by its shape, but by the absence (as pointed out above) of sections of reflector (above and below the lamp), a non-continuous reflector surface, and the expected cut-off and forward baffling. It may have been 'computer designed' but it was obviously not successful.

I'm not convinced by the suggestion that Mk.2 LHD headlights on dip were significantly better. I have a pair of such here (bought when we brought the car to Spain), and they were fitted. The dip beam pattern (if you could call it that) was just dreadful: two very bright spots of light just in front of the car, and gloom to all sides, even with the horizontal cut off correctly set. No even spread across the tarmac just in front of the car, or of the kerb-sides. In the end, I opted for the car's OE Mk.2 units, with non-yellowed fresnels flipped into LHD position.
What are the Morette lights like? Apart from pricey!

Sent from my E5823 using Tapatalk

xmexclusive
Global Moderator
Posts: 5925
Joined: Wed Apr 15, 2009 8:11 am

Re: About to take delivery

Post by xmexclusive » Wed Dec 27, 2017 10:27 pm

I have just been and checked the few headlamp reflectors in the work in progress heap here at home.
All had been removed from recovered units and none were from the new stock.
I examined 7 dip beam reflectors and none showed any sign of heat damage to the silvering or evidence of oxidation.
Six were from Mk 2 lamps and one Mk 1 lamp.
Settled dust was a factor from long service.
Similarly dust was the only factor on the main beam reflectors.
I remain certain that reflector heat damage is not an issue with XM headlamp reflectors.
There are large air vents on all four front sides of the Dip reflector when it is assembled on its extension of the main beam reflector frame.
In service the metal cap on the Dip bulb covers the front half of the bulb and acts as a non reflective light trap and a significant heat sink.
Gap and Cap between them keep much of the generated heat from affecting the plastic reflector.
I do agree careful reflector cleaning after 20 years dust collection is well justified.

As far as the design is concerned as Matthew states there is a massive size difference between the main and dip beam reflectors.
The main beam reflector has roughly three times the surface area compared to the dip reflector.
On the XM both main and dip use computer generated shape complex surface reflectors.
Complex surface is used simply to try to project all the available light evenly into the target area.
The target area is significantly reduced for dip beam than it is for main beam.
The large black areas top and bottom of the dip reflector and the lamp cap are parts of the target area control.

I still consider that Halogen bulbs vary in design quality and also quite quickly reduce actual output in service use.
On the XM the dip bulbs are always running including when main beam is selected.
So in service you are comparing dip beam spread with main AND dip output combined.
Dip will always be seen as very poor in this comparison.
The result of always running is that the dip bulbs age faster than the main ones.
So the advice to use high quality bulbs is important and well known.
Regular bulb replacement at half life well before they reach glow worm state is seldom practiced.
I think this is very important with the dip bulbs.

John

User avatar
Dean
Global Moderator
Posts: 6116
Joined: Wed Apr 15, 2009 7:53 am
Location: Isle of wight

Re: About to take delivery

Post by Dean » Wed Dec 27, 2017 10:39 pm

jonhopper wrote:What are the Morette lights like? Apart from pricey!
Sent from my E5823 using Tapatalk[/quote]

Wonderful!

I think the complex surface light units were just poor and that's it, Rover created a similar shape housing in the 2/4 series cars from the same era and they work really well so I'm not sure how Citroen got it wrong.

The fact we need to talk about cleaning and fresnel yellowing, bulb types and hid just proves the point we are left having to make the best of a bad job imo.

D
92 Citroen XM Prestige 3.0i Auto R.P5678
14 Mitsubishi L200 Trojan
89 Talbot Express 2.0 coach built Auto-trail Chinook

Addicted to Crackanory

xantia_v6
XM Guru
Posts: 1656
Joined: Sun Mar 31, 2013 5:09 pm
Location: Christchurch, New Zealand or Beaune, France

Re: About to take delivery

Post by xantia_v6 » Thu Dec 28, 2017 8:32 am

Regarding the late LHD headlights, as I reported earlier viewtopic.php?f=19&t=4628&start=90#p77088" onclick="window.open(this.href);return false; they are not bad at all.

The only other headlights with internal plastic Fresnel lenses that I am familiar with are the Cibie Biodes fitted to the pre-HE XJ-S, and they also had a reputation for being poor, especially when the lenses yellowed. Jaguar stopped using them by 1980, so there was a lesson that Citroen did not heed. Interestingly the XJ-S headlights have a full size reflector for low beam, with the lens in front of the top 60% of it and the high beam reflector in front of the bottom 40% of the low beam reflector. They also suffer from dust (and corrosion) spoiling the mirrored surface, but they are glued together, so hardly practical to clean, although I did dismantle one and had the reflectors polished and silvered, which made a huge difference.
1999 XM Exclusive V6 24V ES9 Manual (LHD) 105,000 km
1997 Xantia Exclusive V6 (RHD) 45,000 miles

User avatar
Dean
Global Moderator
Posts: 6116
Joined: Wed Apr 15, 2009 7:53 am
Location: Isle of wight

Re: About to take delivery

Post by Dean » Thu Dec 28, 2017 8:39 am

I had a quick search for the Rover 2 series headlamp units and it looks like they were manufactured by Valeo. Did Citroen actually design the headlamps or did they contract Valeo to do it for them?

D
92 Citroen XM Prestige 3.0i Auto R.P5678
14 Mitsubishi L200 Trojan
89 Talbot Express 2.0 coach built Auto-trail Chinook

Addicted to Crackanory

User avatar
White Exec
Citroen God!
Posts: 6642
Joined: Fri Mar 11, 2011 7:38 pm
Orga / RP numbers: RP7165
1996 2.5TD saloon, Exclusive, Polar White
1992 BX19D Millesime, Sable Phenicien
Location: ex-Ealing, Cheshire, W.Sussex & Surrey. Now living in Sayalonga (Malaga, Spain)

Re: About to take delivery

Post by White Exec » Thu Dec 28, 2017 8:53 am

This thread could now be retitled 'Christmas Lights'. A suitably seasonal topic, all the same!

I do agree with John's analysis, and on the point about dirt/dust being the main factors in reducing reflector performance, along with cheap lamp degradation*, and the need for delicate cleaning.

This was also interesting:
" The main beam reflector has roughly three times the surface area compared to the dip reflector.
On the XM both main and dip use computer generated shape complex surface reflectors.
Complex surface is used simply to try to project all the available light evenly into the target area.
The target area is significantly reduced for dip beam than it is for main beam.
The large black areas top and bottom of the dip reflector and the lamp cap are parts of the target area control. "


In the absence of projector-type lenses (wrong era), Citroen/Valeo seemingly chose to 'trim' the dip beam shape by omitting parts of the available reflector area (above and below the lamp), and by masking off light output from the lamp with the metal shield. Both of these non-reflective measures reduce the light output being thrown forward. Only the "side wing" reflectors to the sides of the lamp do anything useful. A good deal of the light going up, down and directly forward from the filament appears lost.

Compared to a typical H4 set-up - where a fully-surrounding reflector is present - the XM arrangement looks to me rather over-optimistic. Maybe the side wing reflectors were very carefully designed, to dispatch light in a top-cut, wide-angle pattern, but at the end of the day they are not able to capture anything like the full output of the lamp.

* Old lamp degradation
My experience is that cheap lamps do indeed lose their output over time more than best-quality ones. On the odd occasions that I have had to replace a single Osram or Philips lamp with a new one, there seems to have been little difference in brightness between the new one and the old one on the other side. With cheap lamps, the difference has sometimes been quite noticeable. Preference is to change in pairs, keeping the good one as an emergency spare.
Chris
1996 XM 2.5TD Exclusive RP7165 Polar White
1992 BX19D Millesime RP5800 Sable
1989 BX19RD Delage Red Deceased; 1998 ZX 1.9D Avantage auto Triton Green Company car 1998..2001; 2001 Xantia 1.8i auto Wicked Red Company car 2001..2003

Post Reply