Film-Tech Forum ARCHIVE: Rectifier Diodes

This is topic Rectifier Diodes in forum Film Handlers' Forum at Film-Tech Forum ARCHIVE.

To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=1;t=003788

Posted by William Phillips (Member # 1162) on 07-31-2002, 03:09 AM:

Whats the best way of checking the efficancy of rectifier diodes and capacitors.
I have just taken out a Osram 1600watt HSC thats done 1750hrs, the picture was starting to flicker.
The cathode is looking a bit pitted. All the other lamps out of my machine look ok.
The lamphouse is a Cinemeccanica CX21H with an IREM N3-X75DM

Posted by Pete Naples (Member # 761) on 07-31-2002, 03:45 AM:

Do you mean how to measure the ripple current?

I use a AC/DC clamp adaptor on my Fluke meter, which makes it very easy. If you buy/beg/borrow/steal one of these make sure it and the meter are true RMS, otherwise the reading isn't especially accurate.

There is a method involving measurement taken across the ammeter shunt, it's detailed on the Osram site somewhere, I'll bet John P can point you in the right direction.

Posted by William Phillips (Member # 1162) on 07-31-2002, 04:36 AM:

Cheers Pete,
I will try and borrow a clamp meter.
I was just woundering if there was a way of checking the diodes to see if they are going to give out before they do, also the cap.
We have had a few very nasty power surges since this machine was in, one of them we had a lot of damage done to equiptment in the building, computers, emerngcy lights, fridges etc. The projector was running at the time. So I would prefer to find out before anything nasty happens.

Posted by John Anastasio (Member # 607) on 07-31-2002, 05:58 AM:

There's really no way to know when a diode is "ready" to go. When it goes, it's a sudden breakdown usually caused by too much inverse voltage applied to it. A diode is a one-way gate, and if you suddenly push too hard against that gate, the latch is going to come loose! It's those nasty surges that do it. They don't exhibit any characteristics that get gradually worse over time. They just blow. Same thing with capacitors. You can't really measure how well they're "aging". The breakdown is caused by a catestrophic failure of the dielectric, the substance that seperates the two plates. When a high enough surge enables an arc to jump from plate to plate, pushing right through the dielectric, it causes a hole. That hole is now a dead short and the cap is fried. What you need to do if you have surges is to either get rid of the surges or to install diodes and capacitors with a higher rating so that the components can take the extra strain.

Posted by Demetris Thoupis (Member # 833) on 07-31-2002, 07:40 AM:

Well,
I had a huge problem with IREM rectifier. Diodes kept blowing too often and I decided to change the whole set with bigger ones. After two years of hard working these diodes have shown no problem. Try that to ensure a safe and reliable projection.
Demetris

Posted by Rick Sanjurjo (Member # 1365) on 07-31-2002, 06:16 PM:

Checking the ripple with a clamp on ammeter is a good method. The other way would be to measure AC voltage across the shunt with your meter and convert that reading to amps. Most shunts will be stamped with a number like 50mv 200a that indicates how to convert your voltage reading to current. In the case of a 50mv 200a shunt you would multiply your reading in millivolts by four and that would be your ripple current. The ripple current divided by the operating current will give you the percent ripple in the power supply.When we check ripple in the factory we use a Tektronix's current probe connected to a scope.

A good way to check the condition of the diodes is to put a temperature probe on them and see how hot they are running. If one is running hotter than the others then there is probably a problem. Another method is to check the voltage drop across the diodes, it should be the same on all of them.

Posted by Rick Long (Member # 213) on 07-31-2002, 10:48 PM:

Diodes are like fuses - they're good until they ain't. Generally, when replacing them, go for the highest current and P.I.V. ratings you can get in the same case size.

At Eaton's Centre, Toronto, I used to have a lot of problems with diodes blowing on the horrible little orange Irems, until I put small blowers on the top of them (blowing upwards). They were attached to the grill, just above the diodes.

Where before, we were lucky if a diode lasted a year, after the fan installation, there were no rectifier calls until the theatre closed, six years later.

Pitting, as you know, is a sign of insufficient current regulation. Might be time for a capacitor change too. Electrolytics do dry out over time.

Posted by Gerard S. Cohen (Member # 994) on 07-31-2002, 11:48 PM:

Hooo--HA! I didn't know you can make a career of clamping rippling
dildos with a Fluke meter and blowing gently on them (upwards)?
I'm certainly glad they dry out by themselves!

Posted by Paul G. Thompson (Member # 655) on 08-01-2002, 12:10 AM:

I remember an IREM that blew a diode at the Keno Theatre in Kenosha, Wisconsin around the year of 1981. No diodes in my kit, and no replacements in the booth. I had a king-sized diode that I clamped on with a vise-grip pliers, and affixed a blower on the power supply.

It kept the theatre from going dark until the next service call.

In short, keep a full complement of diodes in the booth in case one gets blown.

Posted by Pete Naples (Member # 761) on 08-01-2002, 08:41 AM:

My only gripe with using the drop across the shunt is that you have to attach leads to it, which may or may not be easy according the design of the rectifier/lamphouse. Also it's far too easy to leave the meter connected and strike up the lamp, which will almost invariably result in goodbye meter! I've knackered two in ten years that way. My current Fluke is alleged to be good to 16kV overload, but I don't fancy trying it. That said you shouldn't leave the clamp adaptor on whilst striking either, I've found the splash from the striker upsets it too. Safest is to strike up, then clamp around one of the DC leads.

Posted by Mark Gulbrandsen (Member # 72) on 08-01-2002, 12:09 PM:

Indeed ALWAYS replace all of the diodes in a rectifier even if there is only one bad one. Diodes are under tremendous thermal stress while operating and these repeated stresses, after thousands of thermal cycles are what cause many to fail...the junction becomes mechanically unsound due to the epoxy filling deterioating. This epoxy(glass in high quality diodes)eventually begins to loosen up and the junction then fails. Many times you can actually wiggle the stem of the diode after you remove it. Also, its best not to solder directly to the stem of any stud mount diodes. Most are installed with a small screw, lockwasher, and nut, and this is by far the best way to go, especially in the field.
When I was back in the Chicago area I met a retired semiconductor design engineer that really knew his stuff. Some of the above are his reccommendations and some are from my experience over the years.
Most important thing to look at when selecting a replacement diode though is to get one that is glass filled.
Mark G.

Posted by Paul G. Thompson (Member # 655) on 08-01-2002, 01:34 PM:

Mark is correct. Change them all out if one goes goofy. If one is ready to go, they all have the same hours on them, and the other ones will follow. Diodes get leaky, and that can cause some problems.

Posted by Mark Gulbrandsen (Member # 72) on 08-01-2002, 06:53 PM:

I forgot to mention that the really correct way to check them is on a semiconductor curve tracer. There are also most likely more sophisticated computerized tools for doing this now but a curve tracer is still pretty much standard for checking actual specs on each diode(or transistor).
Mark G.

Posted by Rick Sanjurjo (Member # 1365) on 08-01-2002, 10:35 PM:

I agree using a clamp on ammeter is easier and safer than going across the shunt, but if you don't own an ammeter the shunt is an alternative.
Rick

Posted by Rick Long (Member # 213) on 08-02-2002, 10:08 PM:

OK, now how the hell do I tell if a diode is glass filled? Drop it from a great height and see if it rattles?

What is the advantage of a glass filled diode?

Posted by Gordon McLeod (Member # 33) on 08-02-2002, 10:31 PM:

I usually don't replace all the diodes unless I can see major heat related discolouration or know there has been some HV leaking back.
Very rarely have I had a failure after that.
Usually If the diodes are the highest PIV rated available they are ok

Posted by William Phillips (Member # 1162) on 08-03-2002, 05:29 AM:

The diodes in my N3-x75 rectifier are 42HFR100, Does that mean that they are 42amps at 1000volts PIV. There are 6 in total.2 for each phase.

Posted by Gordon McLeod (Member # 33) on 08-03-2002, 10:09 AM:

That is correct

Posted by Steve Guttag (Member # 268) on 08-03-2002, 10:14 AM:

Not entirely correct...42HFR100s are 40 amp not 42. International Rectifier naming convention has the second digit signify the lead. "0" means no lead, "1" means leaded and "2" means leaded with a jacket. Thus your 42HFR100s will be a leaded diode with a red jacket since it is a reverse diode. A forward diode would have a blue jacket.

Steve

------------------
"Old projectionists never die, they just changeover!"

Posted by Will Kutler (Member # 770) on 08-03-2002, 04:32 PM:

Diodes....

Something that no one has mentioned. lamphouses and rectifiers also have many filter caps. If these caps go, rf signals can go back through the system and also cause havoc on diodes! I recently serviced a Strong Super Lume-X and High Reactance Power Supply. Many of the filter caps in the lamphouse were bad and played havoc on diodes in the power supply. (the guys at Strong helped me troubleshoot). They also suggested that I put some filter caps across diodes as extra protection, and these filter caps are not oem, but an after market Strong suggestion.

Soldering diodes can be very tricky, and much care must be used. If you are not an experienced solderer, then stay away! Solder heat will ruin diodes! A generous amount of flux (to be cleaned off afterwords) and a hot, well tinned iron with a excellent tip and heat sinks (generally aligator clips) and proper mechanical connection prior to soldering) is a must!

Also note that many diodes are installed with solderless connectors!

Diodes can sometimes also be defective from the manufacturer.

Ohms scale on a good multi-meter...I personally like analog over digital! can be used to check diodes. Caps need to be checked on a cap tester w/the proper value range!

Posted by Paul G. Thompson (Member # 655) on 08-03-2002, 05:30 PM:

Will, are you referring to the RF suppression capacitors? There are usually several of those in the lamphouses, too. If the power supply diodes seem to fail faster than usual, I would recommend changing those RF suppression capacitors, both in the lamphouse and power supply as well. As we all know, the RF pulse from the igniter will really raise hell with the diodes if it gets on the DC rail.

As far as changing the rectifier diodes, I still change them in sets. Diodes are cheap, and it is very cheap insurance rather than to risk a loss of a show, which usaually happens in a "Full House - Friday Night" situation.

Some Xenon Power supplies (The heavy duty ORC's and some light duty Xetrons) use selenium diodes for the boost circut rectification. Seleniums have such a high internal capacitance that the RF pulse does not bother them too much. So those seldom fail. When they do, they'll really stink up the booth. I do not recommend replacing them with Silicone diodes. I tried that once in a 900 watt IREM, and it simply just did not work for very long. I hung a bunch of suppression capacitors across the diode, and it didn't do the job. I finally broke down and replaced it with an orginal style Selenium rectifier.

Posted by Manny Knowles (Member # 1171) on 08-04-2002, 11:50 PM:

It's interesting that this topic came up because I just had to fly out to another island to change diodes. I don't usually deal with the electrical components, so I sought out instructions from the technician in Miami.

A simple test for continuity in both directions through the diode revealed the culprits (two were bad). I did not have a full complement of diodes, so I wasn't able to change them all. I just ordered a bunch, so that won't be an issue for a while.

I was instructed not to use any chemicals on the heat sink, so I used a guitar pick* to remove the old and also to apply a layer to the face of the new ones.

I was also advised to ascertain a tight connection. (I guess that's why each diode came with a replacement nut and star washer?)

~Manny (The director's cut).

*The guitar picks made it into my tool kit after Bobby Henderson mentioned it elsewhere for cleaning buildup from the film bands in dirty gates. They have come in handy.

Posted by William Phillips (Member # 1162) on 08-08-2002, 12:44 PM:

Ok so i`ve replaced my diodes and caps with the same type.
I know that the diodes are 40amps stud anode with a PIV of 100ov, and the cap is a 4700uF 75v. Putting in better rated diodes like 82HFR100 would be good but how about uprating the cap. I was thinking about maybe around the 10,000uF 100v mark. Would this help to reduce the ripple? or would I be wasting my time, I would love to be able to afford the external filter unit but cant at the moment. The rectifier is IREM N3-X75
Cool

Posted by Mark Gulbrandsen (Member # 72) on 08-08-2002, 02:09 PM:

Paul said "When they do, they'll really stink up the booth."
Ya gotta be real careful ot to breath this in at all. Its extremely toxic! I don't remember what gas it is but its darn dangerous in a big enough dose. John P. is there a link to this gas?

As far as changing the caps to a larger size I'd be very careful you don't increase the inrush current over the amount the lamp you are using can safely tolerate. In the Irem they use a seperate start circuit, but it is possible the larger main caps my increase things a bit too much and eventually blow the anode and cathode to bits. Inrush can be checked with either the ORC Inrush./Ripple meter, or by using a storage scope measuring across a very high wattage precision calibrated resistance...similar to a meter shunt but not actually one. You record, and store the inrush pulse and read it in millivolts dc off the scope, and then calculate it using ohms law. Osram used to have one they'd send out for tests but I don't know if they still do.
Mark

Posted by William Phillips (Member # 1162) on 08-08-2002, 03:00 PM:

Cheers Mark. I think I will leave the caps alone, just do the diodes at some point.
Anyone are the 42HFR100`s that are the correct ones glass filled?

Posted by Antonio Marcheselli (Member # 373) on 08-08-2002, 03:29 PM:

Steve,

perhaps IR has changed their codes... I found 87 HFR 120 and I became crazy to understand what that "7" means. I search the entire IR site and finally, at the end of a PDF datasheet, I found this:

85=Standard Device
86=Not isolated lead
87=Isolated lead with silicone sleeve (Red for reverse polarity, Blue for Normal polarity)
88=Type for rotating application

I can send the pdf if you want!

Bye
Antonio

Posted by Steve Kraus (Member # 476) on 08-08-2002, 05:31 PM:

Is there any value to replacing all the big caps just on principle on old rectifiers (+20 years) or is it better just to wait for things to fail?

Posted by Steve Guttag (Member # 268) on 08-08-2002, 09:15 PM:

Nope, they haven't changed their codes...I should have stated that it is +0, +1, +2 to the current rating...thus an 87 is an 85amp diode with a covered lead (red for reverse, blue for forward).

Steve

------------------
"Old projectionists never die, they just changeover!"

Posted by Rick Sanjurjo (Member # 1365) on 08-09-2002, 07:58 AM:

I wouldn't increase the size of the filter caps unless I could measure the inrush current. Excessive inrush can cause short bulb life. You can check the ripple on your supply and if it seems high then replace the caps with the same value ones. The job of the capacitors in the supply are two fold, one they provide the right amount of inrush current to get the bulb lit and two they filter the DC from the bridge to lower the ripple current.

Rick

Posted by John Pytlak (Member # 331) on 08-09-2002, 09:34 AM:

Rick: What is the typical capacitor failure mode as they age (e.g., short, open, loses capacity, leakage)? Would there be some way of checking the capacitors by charging them to a known voltage, then looking at the R/C characteristic as they discharged through a known load? What is the typical value used, and is it different for different types of rectifiers (e.g., full wave bridge, 3-phase full wave) or lamp types?

------------------
John P. Pytlak, Senior Technical Specialist
Worldwide Technical Services, Entertainment Imaging
Research Labs, Building 69, Room 7525A
Rochester, New York, 14650-1922 USA
Tel: +1 585 477 5325 Cell: +1 585 781 4036 Fax: +1 585 722 7243
e-mail: john.pytlak@kodak.com
Web site: http://www.kodak.com/go/motion

Posted by Gordon McLeod (Member # 33) on 08-09-2002, 02:28 PM:

One problem with capacitors is that they also have a ampre rating
In otherwords how much current will they be charging and dischargeing over time
Often the smaller caps even though the Voltage and capacity ratings are correct they will have trouble and overhead with current

Also if the rectifier has a filter choke in it I usually move the caps to the input side of the choke and then add a small one on the output to make it a P filter
The small one helps the ignition but the big one being before the choke have ther discharge time constant dampped

Posted by William Phillips (Member # 1162) on 08-09-2002, 02:33 PM:

Thanks for all the replays,
Has any one out there got one of the IREM AFU external filters, if so are they any good

Posted by Rick Sanjurjo (Member # 1365) on 08-09-2002, 05:40 PM:

John,
I have seen them go both open and short. The short circuit condition is easy to find, there is usually smoke. The open condition is a little harder the symptoms might include short bulb life and hard to light bulbs. The exact value is dependent on the power supply design. For example a single-phase supply would have higher ripple current and require larger capacitors as compared to a three-phase supply. A supply with a higher no load voltage would require less capacitance than one with a lower no load voltage. Some supplies like the Strong high reactance insert a limiting resistor in series with the caps during ignition so that they can use larger value caps during normal operation (for lower ripple) and still not get excessive inrush at start up. It is a tricky question as to what the correct value for a given supply should be. If you want to check the capacitors most modern digital meters include a capacitance meter. If not you could charge them to a known voltage and discharge them through a known resistor and measure how long that takes and then calculate how long it should take, but that’s kind of a hassle.

Rick

Posted by Paul G. Thompson (Member # 655) on 08-10-2002, 02:43 AM:

One thing I will add that Rick didn't. An open diode will give a symtom of low bulb current (substantially lower, I might add) accompanied with shutter strobing on the screen.

I don't like using capacitor checkers. They don't tell the full story.

Looking back on electronic theory I seem to remember that if a capacitor checker is used to check capacitance and the results are higher than expected, the capacitor is electrically leaking.

Can someone refresh my memory on that?

Posted by John Walsh (Member # 168) on 08-11-2002, 03:27 PM:

I wrote a procedure to check Irem rectifiers, which I did on a yearly basis (I wish I had that kind of time today!) A person might be able to apply it to other makes.

Several people at Amprobe and Irem's US reps told me that using an AC clamp on amp meter would not be too accurate. Several people would try to use their AC clamp-on meter, since ripple is a form a of AC.

People might take issue with my suggesting to disconnect the AC ground to the 'scope. Battery-operated 'scopes were mostly dream when I wrote this. So, use at your own risk, and let common sense prevail, but I still do it this way today.

Determining the percentage of ripple in
power supplies for short-arc xenon lamphouses.

This procedure also checks for any error in the voltage and/
or ammeter.

Before beginning, verify that the power supply is operating at
or below the AMPERAGE shown on the power supply nameplate.

1. With a small flat head screwdriver, "zero" the volt meter and
ammeter (if installed).

2. Set the oscilloscope to 10 millivolts per division; the time
base to 20 milliseconds per division.
Avoid using "X10" probes.
There are no special triggering requirements, AC or "AUTO"
will work.

Note: Since the oscilloscope will be reading DC, greater
accuracy can be obtained by adjusting the scope's
baseline to the bottom of the screen, rather then to
the center.

3. Place the auto-ignition switch to "MANUAL".

4. Open the lamphouse door and defeat the door interlock safety
switch.

5. Observing proper polarity, connect a calibrated volt meter to
the DC power cables.

6. Power on the lamphouse. The bulb should not ignite while the
auto-ignition switch is in "MANUAL".

7. Compare the voltage reading from each volt meter. Record the
values. Replace the meter in the lamphouse if the values are
more then 15% apart.

8. Remove the power to the lamphouse while the meter is still
connected.

9. While observing the voltage decrease, listen for the RA
relay (inside the rectifier) to drop out. Record this value.
It should be between 30 to 45 volts.

Note: It should take longer then 10 seconds for the meter to
reach zero.

10. Remove the volt meter when the value reaches zero.

11. Power on the lamphouse.
a. Press the "STRIKE" button to ignite the lamp OR:
b. If there is no "STRIKE" button:
Flip the auto-ignition switch to "AUTO", then
immediately return it to "MANUAL" after the bulb is
lit.

The auto-igniter must be disabled. If the bulb should go
out, it can not re-ignite and injure someone
connecting leads or damage test equipment.

12. Connect the oscilloscope across the ammeter shunt.
Record Imax (total peak voltage)
Record Imin (minimum peak voltage)
Calculate ripple and record. The percentage of ripple should
be less then 10%.

Note that the oscilloscope may have to be isolated by plugging
a ground adapter in series with the oscilloscope's power line.
Use caution.

13. Record the lamphouse ammeter value.
Because the ammeter shunt reduces the actual current
value to the lamphouse ammeter by 1/3000, there
will be 3 amps for every 1 millivolt shown on the
oscilloscope. Therefore, the amperage can be calculated
by multiplying Imax, in millivolts, times 3000.
For example:

36 mv * 3000 = 108 amps

Note that this test requires an accurate 150 mv shunt.

14. Compare the readings from the ammeter and the calculated
current. Record the values. Replace the meter in the
lamphouse if the values are more then 15% apart.

15. Return the door safety interlock switch to normal.

The following information should have been recorded:
The date of testing.
The lamphouse model, serial number, and location.
The power supply model and serial number.
The values: Imax; Imin; operating lamphouse voltage; actual
voltage; operating amperage; actual amperage; RA relay drop-
out voltage; percentage of ripple.

Posted by Steve Guttag (Member # 268) on 08-11-2002, 03:54 PM:

Or...

Just pitch the IREM altogether and install your favorite other brand and watch the lamp life increase and realize, it was probably the ripple in the beautiful to look at but horrible power IREM rectifier.

Christie rectifiers powering Christie lamps...has got to be the toughest act to beat. To consistantly obtain 2-3 times rated life out of lamps, regardless of size speaks to their construction.

Steve

------------------
"Old projectionists never die, they just changeover!"

Posted by Mark Gulbrandsen (Member # 72) on 08-11-2002, 09:24 PM:

Steve,
I have to adamantly disagree with you on the Christie rectifier thing. They are sooooooooooooooooooooooo old fashioned that its almost a joke. There are other brute force rectifiers I'd use over a Christie any day. Man, I've had less problems with Eprad rectifiers than Christies. I do however agree with you on the majority of Irems though. They put out dirty DC,typically 9 to 10% ripple on a new unit, and have enough of their own problems to fill the Titanic. However, Christies are way under heatsinked, use crappy cheapo IR epoxy filled diodes that WILL fail in alot less time than others that use glass sealed type. The crappy red fibreglass board that all the high power connections, in fact all the connections period attach to compresses and allows things to loosen up over time and connections to get hot and burn up. About as good as Kneisley there! I can't begin to tell you how many Christie rectifiers I have done service calls on in 20 years. WAY more than any other single brand indeed. These days switchers are in and provide the best source for a lamp one can get. They are super reliable and don't allow the current to drop during electrical brownouts(within reason). Sure, they have the one downside that they usually cannot be repaired in the field, but it only takes 5 min to change one out, alot less time than it takes to change diodes in a Christie. Gee, on the SLC consoles Christie finally put a fan on the heatsink..... Its about time. They went for years with out them in the H series.

JOhn, Osram use to have a standard .1% accuracy shunt they would send out for investigating lamp failures. Your procedure is very correct except I used a storage scope instead and also measured and calculated inrush current at the same time, using the same shunt. I ahev an ORC ripple and inrush meter that I ahve had for years. Is till use it fairly often and ahve compared it to the standard shunt on one occasion. The amprobe they use is an AC/DC version and it feeds a special circuit in a black box. The calibration of the ORC unit was pretty close to the standard shunt so I always just stuck with the ORC system.

Mark

Posted by Steve Guttag (Member # 268) on 08-11-2002, 09:40 PM:

You can disagree all you want and critique the age of the Christie design...it still works best...it will make any lamp last longer.

So basically, what you are saying is to use glass filled diodes and everything will be fine on the diode failure front? Hmm. Seems like a cheap solution to me.

The current Christie standalones all have fans on them and Christie was putting fans on their console units about halfway into the K series.

As to switchers vs Christie...even Cineplex pulled the switchers out for the Christies at the Uptown after the 6th failure...the result, no rectifier failures in years and much improved lamp life.

Mark, you know that newer isn't necessarily better...it isn't an excuse to use something...what switchers bring to the table is size and weight. I do use them for smaller lamps...look at my picture in the picture test thread..there is a Strong switcher sitting nicely between the Kinoton's feet, like it was sized to go there!

Steve

------------------
"Old projectionists never die, they just changeover!"

Posted by Mark Gulbrandsen (Member # 72) on 08-12-2002, 11:35 AM:

Humm, I never had a customer that had a lamp last any longer with a Christie rectifier than a switcher. My experience has been that they last about the same. The switchers have SO many advantages that its almost difficult to think of anything else for mainstream installation. And the quick changeout is a BIG advantage. Yes, on special high power setups they can be problematic but even there have been pretty darn good for me. Anything over a 7kw should see DC from a modified Miller Arc Welder Power Supply anyway. For really large lamps thats the best way to go.

As far as diodes go always get replacements thata re glass filled. Its easy to spot an IR cheapo epoxy filled version. Dull looking blue or green colored fill. Glass filled/sealed diodes have a very shiny appearance to them. My experisnce is that if you use glass filled you almost never ever go back, unless lightening hits.
Mark

Posted by John Walsh (Member # 168) on 08-12-2002, 09:10 PM:

The equipment I service is almost all Xetron consoles with IREM power supplies. We usually get 3 times the rated hours on a bulb; often I will (or I'll suggest) they be changed just because there's so many hours on them I think they might blow up. I think IREMs are very reliable. YMMV, but I find airflow to be much more important than the make of power supply.

In our entire chain, we've changed very few diodes. But for some reason, I've had to change 6 within the last two months; At two separate theaters the same thing happened; two went in the same power supply, then one in the house next to it. Each one went a few weeks apart from another. Very strange. At our Miami FL location, the power is always going out, brownouts, etc. but no diodes have blown. In our Trumbull CT location where the power is well regulated, three have gone bad in the last 2 months. So, go figure.

I also see no "logic" in changing out all diodes if just one is bad (unless, as Gordon pointed out, there is a specific reason.) IREMs are convection cooled, so I thought they might be overheating since they are now being placed in an enclosed console. But many older consoles are sill working fine, so I don't know.

Posted by Steve Guttag (Member # 268) on 08-12-2002, 10:05 PM:

Mark,

Perhaps your customers don't see it but I have repeatedly. As to IREM, they consistently rack up the shortest hours with the most ripple.

Steve

------------------
"Old projectionists never die, they just changeover!"

Posted by Mark Gulbrandsen (Member # 72) on 08-12-2002, 10:38 PM:

Steve, typically Irems run from 9 to 12% ripple. The inrush specs are usually very good though.
Mark

Posted by Mark Hathaway (Member # 1044) on 08-14-2002, 04:27 AM:

It would be best to measure ripple current with an oscillscope across the shunt, as even a good multimeter will probably struggle resolving the waveform.

Mark Hathaway
Senior Technician,
Atlab Image and Sound Technology,
Melbourne, Australia

Posted by Pete Naples (Member # 761) on 08-14-2002, 10:51 AM:

This is fascinating! I've never had a ripple problem with an IREM, if any rectifier was giving me 9-10% from new it'd be back in it's crate on the way to meet it's maker! The IREMs I service typically give 3% at the outside. I wander if this is another 50/60Hz thing?

Now the Christies.... those seem to be pretty good too, the 12 that are local to me are knocking on in years a bit now, but have been fretty reliable. All recently had new smoothing caps fitted as the ripple was getting a tad high. I had one real spectacular failure on one Christie. Two diodes failesd short, and in turn took out most of the snubber network in smoke and flames fashion. I had to completely remove that rectifier from the console to work on it and rapidly discovered that the Christie is surprisingly heavy for it's size!

Posted by John Walsh (Member # 168) on 08-14-2002, 06:10 PM:

I'm not totally convinced that ripple plays a large part in determining bulb life. I'm not saying I couldn't be convinced.....
Obviously, if it was like 40%, well sure.

I measured several electrical parameters of IREM rectifiers, ripple being one of them. Now, I gathered this data about 15 years ago, but, at least at that time, Steve was right. The 15 IREM's power supplies I measured had ripple between 9 and 18%, only one was really low. But we didn't have any bulb problems, so as I say, I don't know about bulb life vs. ripple.

Posted by Richard Fowler (Member # 893) on 08-14-2002, 09:10 PM:

Excessive ripple will darken the bulb even if you have "rated" life.
Richard Fowler
TVP-Theatre & Video Products Inc. www.tvpmiami.com

Posted by John Pytlak (Member # 331) on 08-14-2002, 09:21 PM:

I guess with current ripple, those tungsten ions are sometimes going the wrong way on what should be a one way street.

Strong Xenon Article
LTI Article
CES Info Page

Posted by Paul G. Thompson (Member # 655) on 08-15-2002, 01:10 AM:

Mark and Steve:

BAH! I think Christie rectifiers used in their newer consoles are the greatest thing since sliced bread! The bulb life I got out of those things approach 6000 hours on a consistant basis! Must be something good about it.

OK - flame me.

Posted by John Walsh (Member # 168) on 08-25-2002, 10:37 AM:

Sorry to drag this topic up again, but I thought I'd take new readings and see what the ripple was on our current IREM supplies. As far as ripple in IREM's, it looks like things haven't gotten any better. In the IREM documentation, they generally claim their supplies will have no more than %9 ripple, but we're way past that on most units. Pete N. had an interesting point that maybe the power cycle rate (50hz vs. 60hz in the US) might be a factor, but I would think that the US's higher rate would help reduce the ripple.

Before I condemn them though, (and start trying to improve some of the worst ones) I like to be sure I'm doing the measuring correctly. I think my scope is OK, but I'm not sure about the shunts in the consoles. Does anyone know where I can buy an accurate shunt? I can't very well go to ORC and say: "Can I use your shunt to test Xetron lamps?"

Although, even if my scope was out of calabration and the absolute values are wrong, (Imax vs. Imin) the readings are relative to each other. So, I sort of have to conclude the ripple is as bad as I'm seeing.

The data:

Aud IREM Year Imax Imin Ripple

Aud 1 G3X-100 1993 32 27 15.63
Aud 2 G3X-100 1993 33 28 15.15
Aud 3 G3X-100 1993 31 26 16.13
Aud 4 G3X-100 1993 31 27 12.90
Aud 5 D3X-141 1983 36 31 13.89

Aud 7 G3X-100 1993 31 27 12.90
Aud 8 G3X-100 1993 32 29 9.38
Aud 9 G3X-100 1993 30 27 10.00
Aud 10 G3X-100 1993 32 29 9.38
Aud 11 G3X-75 2000 23 20 13.04
Aud 12 G3X-75 2000 25 22 12.00
Aud 14 G3X-131 2000 32 29 9.38
Aud 15 G3X-131 2000 32 28 12.50

At another theater:
Aud 1 G3X-100 1995 32.5 29 10.77
Aud 2 G3X-100 1995 32 30 6.25
Aud 3 G3X-100 1995 29 27 6.90
Aud 4 G3X-100 1995 27 25.5 5.56
Aud 5 G3X-100 1995 32 25 21.88
Aud 6 G3X-100 1995 32 29 9.38
Aud 7 G3X-100 1995 33 30 9.09
Aud 8 G3X-100 1995 24 21 12.50

IREM = IREM model number
Year = year purchased
Imax / Imin is in millvolts,
measured across the shunt w/ Tex 465 scope

ripple = (Imax - Imin) / Imax * 100