Durst Laborator; problems with EST 450 unit.

[Marco Gilardetti]

Good morning everybody.

Unfortunately my Durst Laborator is having bad problems. It burned out three replacement lamps in few minutes. I suppose the problem lays with the power supply unit EST 450, probably the stabiliser isn't working properly.

The very strange thing is, however, that the lamps seem to be UNDERpowered. When I replaced them, everything seemed all right, perhaps just a bit too dim, at the beginning, but within a minute or so the light began getting dimmer and dimmer until the lamp failed and showed a glass all covered by a grey-black powder. Also, I measured the voltage fed to the lamp with a high quality tester and, without load, it read 21 Volts AC rather than the expected 24.

Does anyone know if these 24V 250W halogen lamps can fail for undervoltage? Perhaps by poisoning the gas inside the glass bulb, or by ionisation or such? It's such a strange thing.

Also, does anyone experimented something similar and can guess which component inside the power supply can be faulty? It's quite a complicated circuit, very hard to figure out, but none of the components shows obvious traces of damage.

The head fan is working properly, so I would exclude a problem with overheating. Lamp failure also happens too fast to be due to overheating inside the head.

Thanks to anyone who will help to find out what the problem is.

 

[Marco B]

According to one site I saw referencing the required ELC 24V/250W lamps, the official life-time of these bulbs is just 50(!) hours. Probably not a big deal for an enlarger switching on just seconds each time used, but it might explain why lamps may fail almost instantly like in your case if the power supply has a problem. These lamps seem to be pushed to the limits in usage.

I doubt if "under-voltage" could be an issue. It is a halogen lamp, which is equivalent to a incandescent lamp with a hot metal filament producing the light. The halogen lamp contains an inert gas, which can't react chemically with anything. It is the filament which "burned" up in those minutes, causing the apparent "soot" on the inside of the bulb.

Gas discharge lamps, like a fluorescent energy saving lamp, could have issues with changes in voltages, as they need to maintain an "arc" of excited gas ions (think of a kind of controlled thunder lightning - the arc produces the light!) in the gas discharge tube, leading to a possible premature death of the lamp in cases with power issues.

 

[vorgonyi]

Rüdiger Mehlmann

have you tried asking Mr. Mehlmann.


Mehlmann Fototechnik
--- autorisierter Durst Kundendienst ----
Schillerstr. 9a
D 74889 Sinsheim
RMehlmann@durst-image.de

Telf. +49 (0) 7261 17506
Mobil +49 (0) 172 6313316

he was very helpful with a problem I had.

best

 

[GeorgesGiralt]

Hello !
Halogen bulbs can die from undervoltage/power. This is due to the fact that the tungsten filament vaporize in normal use (slowly, I have to admit) and deposit itself on the quartz envelope of the bulb. The halogen inside the bulb react with the deposited tungsten and return it to the filament. This occurs only at the specified temperature of the filament which is greatly impacted by the voltage the lamp gets. this improve bulb life expectancy and output quality (as the bulb does not get covered with a thin layer of non transparent metal with age).
As these bulbs are quite expensive, I would test run the power supply with a resistor in place of the bulb (a household iron or something like that comes to mind) because the stabilization process needs to supply the correct voltage and power to work.
In other words, measuring the voltage without load is irrelevant as the regulation is not meant to work so could be dead and your meter not seeing it. If you force the current to reach the value the lamps normally draws you will see if the regulation works fine or is dead/havoc.
I do not know this particular circuit but these devices normally limit the power and regulate it at the 250 W specified and, of course, at the nominal voltage to prevent color temperature change in the bulb's output.
Bear in mind that this electronics is from the seventies and could be impossible to repair due to the lack of components which can be obsolete... So do not make destructive testing ....

 

[Marco B]

Hello !
Halogen bulbs can die from undervoltage/power. This is due to the fact that the tungsten filament vaporize in normal use (slowly, I have to admit) and deposit itself on the quartz envelope of the bulb. The halogen inside the bulb react with the deposited tungsten and return it to the filament. This occurs only at the specified temperature of the filament which is greatly impacted by the voltage the lamp gets. this improve bulb life expectancy and output quality (as the bulb does not get covered with a thin layer of non transparent metal with age).

I knew about the re-deposition of tungsten back on the filament in halogen, but never realized it was so much influenced by the temperature of the lamp, as you always see dimmers on halogen units. I never realized they would actually potentially negatively affect lamp life.

Ah well, I now see that Wikipedia states:

"Tungsten halogen lamps behave in a similar manner to other incandescent lamps when run on a different voltage. However the light output is reported as proportional to V3, and the efficacy proportional to V1.3.[13] The normal relationship regarding the lifetime is that it is proportional to V − 14. For example, a bulb operated at 5% higher than its design voltage would produce about 15% more light, and the efficacy would be about 6.5% higher, but would be expected to have only half the rated life."

So it seems lower voltage still expands life expectancy of most halogen bulbs... but it depends, another page says this:

"Will dimming my halogen bulb extend it's life?

The halogen cycle can be interrupted if the voltage and temperature of the lamp is operated too low. The result is lamp blackening. Lamp blackening and the consequent loss of lumens are standard for incandescent lamps, but halogen lamps are designed not to blacken as they age. Halogen lamps maintain their lumen output throughout their lifetime due to the key mechanism of the halogen cycle which binds with the tungsten that is vaporized off of the lamp filament itself. Thus keeping the bulb walls clean. If lamp engineers could figure a way to get the tungsten to redeposit evenly back on the lamp filament, we could have a lamp that lasts forever. But, this is not the case, as the tungsten is redeposited on the coolest part of the filament (usually the ends of the filament), and consequently over time, the middle portion of the filament eventually thins, forming a weak spot, and eventually the filament breaks.
The standard operating temperature of the bulb wall to maintain a halogen cycle is 250C. When the temperature falls below this level the halogen cycle fails, and the lamp starts to act like a standard incandescent. It is estimated that the cycle fails when a lamp is operated below 70-80% of its rated power."

So, at some reduced voltage, you probably do get reduced life span, as the halogen cycle breaks down.

 

[Marco B]

And another nice page on this subject of halogen cycle is:

Halogen Regenerative Cycle:
http://zeiss-campus.magnet.fsu.edu/tutorials/halogencycle/index.html

Marco

 

[Bob-D659]

Measuring the voltage without a bulb in place may be an indication of a problem. Some stabilizer circuits don't work very well without a load and will hold the voltage below normal.

Check for loose wires, poor contacts, including the bulb socket itself. The socket should almost be replaced when you change a bulb, as the spring contacts inside it go thru many heat cool/cycles in an enlarger and that weakens the springs.

Good bulbs don't fail in a couple of minutes from running 10-15% under voltage, they will last for days. Sounds like all your spare bulbs came from the same batch and were all bad to start with, possibly due to damage in shipping. I would try to return them for a refund.

 

[ic-racer]

1)None of the TRA450 schematics I am aware of show a stabilization circuit. Basically its just a 24v transformer and some additional 'housekeeping' circuitry.
2) Under voltage will not cause that bulb to blow. As an example I ran the 18 V EKG bulb from an enlarger off a half-wave rectified 13V circuit for 20 years with no problem.
3) Bob-D659 gives good advice on the socket. I have taken the rivets out of sockets and taken them apart to restore the contacts but new sockets ARE available if you look around. The names/numbers of the sockets don't always match the bulb. For example the socket for the Omega D5500 is called QLV-1 but the bulb is a MR-16 style.

 

[Marco B]

2) Under voltage will not cause that bulb to blow. As an example I ran the 18 V EKG bulb from an enlarger off a half-wave rectified 13V circuit for 20 years with no problem.

Yes, I guess we wouldn't be running all those nice halogen dimmed lighting options in houses otherwise, if they broke down each time you lowered the light output to prepare for a cosy evening with your peer. Still, that concept of the halogen cycle and its chemistry is nice to learn about

3) Bob-D659 gives good advice on the socket. I have taken the rivets out of sockets and taken them apart to restore the contacts but new sockets ARE available if you look around. The names/numbers of the sockets don't always match the bulb. For example the socket for the Omega D5500 is called QLV-1 but the bulb is a MR-16 style.

Is it than because the socket due to repeated heating/cooling cycle, simply no longer makes proper contact, and causes unwanted arcing / charges to jump over to the lamp contacts, resulting in some form of damage? What is the cause of the failure of the lamps in this case? Shouldn't one see clearly irregular lighting in this case and is it consistent with the description of the way of failure the OP desribed?:

"The very strange thing is, however, that the lamps seem to be UNDERpowered. When I replaced them, everything seemed all right, perhaps just a bit too dim, at the beginning, but within a minute or so the light began getting dimmer and dimmer until the lamp failed and showed a glass all covered by a grey-black powder. Also, I measured the voltage fed to the lamp with a high quality tester and, without load, it read 21 Volts AC rather than the expected 24."

Or is there still the option of some power supply issue?

Marco

 

[36cm2]

1)Basically its just a 24v transformer and some additional 'housekeeping' circuitry.

Not to hijack the thread, but I believe my Durst L1200 stabiliser was fried last week due to a power surge. Does this mean I can just replace it with a generic 24v transformer, or am I oversimplifying?

Thanks,
Leo

 

[Bob-D659]

Stabilizers, stabilized power supplies and power supplies like the TRA 450 are all different beasts. All depends on what functions your L1200 stabilizer provides. A generic 24 volt transformer is unlikely to deliver 10-15 amps needed for a high wattage bulb.

Open up the cover and post a pic or three in a new thread and someone may be able to help.

 

[ic-racer]

Not to hijack the thread, but I believe my Durst L1200 stabiliser was fried last week due to a power surge. Does this mean I can just replace it with a generic 24v transformer, or am I oversimplifying?

Thanks,
Leo

A few different comments here.

1) does your unit have the EST500?? If so, there is Dead Link Removed (to keep spurious signals from bothering other AC equipment on your lines) that has a tendency to fry in those units. Replacing the capacitor should fix it. See "PME 271" in photo below (J. Bannow photo):

2) A voltage spike big enough to fry the transformer would be quite wild. I'd think you'd have arcing in the wires all through your house if that were the case.

3) A generic transformer with the same Watt (or VA) rating as the original should work fine to power your bulb. These transformers can be expensive, though.

 

[ic-racer]

Is it than because the socket due to repeated heating/cooling cycle, simply no longer makes proper contact, and causes unwanted arcing / charges to jump over to the lamp contacts, resulting in some form of damage?

I'm not 100% sure (based on the EST450 schematic) how a bad socket can fry the bulb in the original posters case. I have good evidence that in TRIAC based regulated supplies (where a TRIAC is the only thing between an 82V bulb and the 120V mains) oxidation (from arcing) in the socket leads to high resistance, causing the TRIAC to open up to "full on" which increases the voltage up to 120V and a spark will jump from the bad socket contact to the bulb pin and instantly fry the bulb. In this case there will be an identifiable 'weld mark' on the bulbs pin. (Ask me how I know, I fried 3 in a row in about 10 minutes before I figured it out one day). The more this happens, the worse the socket gets, because the arcing leaves non-conductive carbon residue each time it happens.

 

[Bob-D659]

As a WAG, I'd suspect the socket with the EST 450 or any supply, as if it's arcing badly it will overheat the pin on the bulb and fracture the seal in a few minutes and air will kill the filament.

 

[GeorgesGiralt]

I forgot to add that I fried a bulb in the TRA450/CLS450 head I own because one of the big wire in the plug at the end of the head cable was unsoldered and made intermittent contact. I think you should have a look at the plug and sockets !

 

[Marco Gilardetti]

Thanks for all the insights and suggestions, I appreciate it.

1)None of the TRA450 schematics I am aware of show a stabilization circuit. Basically its just a 24v transformer and some additional 'housekeeping' circuitry.

That's allright. However, I own a EST 450, which is clearly stabilised, with a big big power transistor mounted on a big big heat sink. It also has a heating simulator made with a power resistor and a heat sensor which turns the head fan on and off automatically. Everything (fan, lamp, timer) is switched by relais. A very sophisticated machine.

TRA = TRAnsformer
EST= Electronically STabilised

The lamp socket is made of first-choice porcelain with heavy-duty contacts and firesafe cables, all in perfect shape and constantly fan-cooled. I would definitely rule it out.

It must be said that the lamps used in the CLS 450 head have almost nothing to do with those of the households appliances. "Normal" halogen lamps cost one buck or less and are usually made in China. Those for the CLS 450 head are made exclusively in the U.S.A. by General Electric and cost € 30 each, to say one thing. Also, they are "pulled" to the extreme, as Marco B already said. Thus I suppose everything, good and bad things, happen much FASTER than in "normal" halogens. So the symptoms are fully compatible with the theory of insufficient voltage: the glass gets progressively darker and darker because of a grey-matter deposit (Tungsten, most probably), until the filament brakes. The whole cycle takes 3-4 minutes.

I suppose the voltage comparator or the voltage reference or both are faulty, as the output power transistor look allright and shows no sign of stress. It's hard, however, to draw things out without a schematic.

The lamps themselves are all right, as the first which broke yesterday had been previously used for many weeks with no issues at all, and the replacement all came from its same batch.

@ Vorgonyi, thank you very much for that address; I tried to contact Durst Italy (no reply so far) and it case that attempt fails I will try to ask Mr. Mehlmann. Thanks a lot, I appreciate it.

 

[Bob-D659]

I have a CLS450 head and the early model TRA450 power supply, the lamp socket has been replaced at least twice, once by me and at least once before I bought it. They do wear out, and from the outside they still look perfect. The bulbs used in the CLS450 are standard projector bulbs and are also made in Mexico and Japan as well. Probably in China too.

If your EST450 uses a relay to switch power to the bulb, check the contacts in it, they get pitted and burned and cause all sorts of problems.

 

[GeorgesGiralt]

And triple check the plugs at the end of the head's cable, plugging into the EST450. If one wire is making bad contact, it cause nasty things to the power supply and the bulb...

 

[Bob-D659]

Just adding, both your EST450 and the CLS450 were made from 1974 to 1979, so both are at least 30 years old, and Durst stopped support and supplying parts in 1989. Thousands of heat/cool cycles cause items to wear out, including the solder joints as GeorgesGiralt pointed out. When you use a 250 watt bulb that draws 10 amps of current, it is recommended to change the bulb socket for every second bulb you replace. Durst USA have a replacement power supply, it is approx $1,000. The earliest TRA450, the one with a two pin european type socket on the back is the easiest one to keep going. It has the fewest parts and only one high current relay to replace when the contacts on it burn out.

Link to Durst products and support dates: http://www.durst.it/fileadmin/user_upload/Durst/pdf/Durst_Equipment.pdf

 

[Marco Gilardetti]

Don't say it loud but I think I've finally fixed the issue. I have to thank once again fellow Georges Giralt for sending me the EST 450 tech manual complete with schematics, that although written in mixed french/german has been extremely useful in understanding this complicated machine.

With reference to the schematics, I replaced all electrolytic capacitors C1 C2 C3 and C5. Especially C3, although aesthetically in perfect shape, was totally out of order and was most probably the main cause of the issue.

Also resistors R42 and R43, which showed traces of burns (extended to the PCB), have been replaced. The power resistor R42 undergoes a very heavy duty and I decided to overrate it at 10W; still it heats up considerably so it was previously underrated in my opinion. R43 was instead probably OK (maybe its traces of burn came up from R42 below) but I replaced it nonetheless while on the job.

The head fan thermal sensor S3 has also been replaced; although it could not be the cause of the general out of order it behave erratically in the last months and clearly asked for being replaced.

I tried the enlarger for 3, 5 and then 10 minutes of continuous operation, surveying the light intensity with a lightmeter, and everything seemed all right. Now the final test is needed: a printing session, that I will do as soon as possible.

As a side note, in case anyone will experience problems in the future: it is normal that the tension output to the lamp reads below 24V, it was done intentionally by Durst to prolong lamp's life. However, the voltage is not a sine wave, so it can be properly measured only at the oscilloscope.

Also, on the center of the PCB there's a small metallic cube which contains a light sensitive resistor and a micro lamp that simulates the headlamp behaviour. It is important to check that this small lamps is not burned out. It can be tested with a 9V battery by removing the black paint from its back: if in order, its weak light should be visible in transparency.

 

[Rombo]

I had a problem with lamp on my Durst Pictochrom. After replacement of burned lamp, I burn three more. Shutter did not opened, and that was the cause. Sensor did not recieve enough light (because shutter did not open).
I do not know, if Laborator 1200 have shutter. If does, check it.