D5500 Power Supply--Complete Rebuild From Scratch

[ic-racer]

I have been planning on doing this for some time but finally am getting around to it.

Project: Completely rebuild a D5500 power supply module.

History: I have 4 power modules that work with varying degrees of competency. One blows bulbs occasionally and another works except the light won't go on. Another works but the filters 'hunt' a little too much.

The service manual recommends "Power Supply Board Replacement" for all the above conditions. F-point used to carry Omega parts and had a brand new power supply board listed for $200 USD for many years. I was going to buy it as a spare but it turns out they got out of the parts business. (I sure hope that power supply did not wind up in the trash. )

There really are not that many components on the power supply board and it looks like one could just replace all components to re-manufacture a brand-new board.

Component Selection:

Even though D5500s were available new into the late 90s, it appears they were assembled from batches of components that were acquired when the enlarger was first introduced in the mid 80s. So, most all of the semiconductors in the power supply are outdated.

I used the D5500 shop manual parts list and the schematic, along with detailed visual analysis of each part. Paying attention to matching operating characteristic and mounting hole pitch and component size.

After a lot of internet searching and analyzing all the component datasheets I came up with a list of every component on the power board except the two transformers and the ribbon connectors (which I plan to leave in place). Everything else has been sourced and will be replaced.

 

[ic-racer]

These 4 jpgs contain a list of all the components from the D5500 power supply board with cross-reference labels from the D5500 parts list and current Mouser.com or Digikey.com part numbers.

 

[ic-racer]

I have 3 complete working head/controller systems and 4 power modules. I plan on re-building one of the power modules and if all goes well, to do a second one or perhaps a third also.

The D5500 controllers have about twice as many parts, including some ePROMs that are unique to the D5500. So, a similar re-building approach for the controllers is not practical. Fortunately, I have 3 controllers that seem to be in good working condition. Since the controllers can easily be swapped from one enlarger to the next, only one working controller is needed anyway. Whereas it takes from 15 to 20 minutes to swap out the power module.

 

[ic-racer]

Here is a picture of a D5500 power board, ready to be re-built.

 

[epatsellis]

Man, you're really going about this the hard way. I would start by measuring all the voltages with reference to the schematic. quite likely you have a single or at most a few resistors or caps that have gone off value. While I admire your dedication, it's a rather simple repair that shouldn't be all that hard.

 

[L Gebhardt]

Man, you're really going about this the hard way. I would start by measuring all the voltages with reference to the schematic. quite likely you have a single or at most a few resistors or caps that have gone off value. While I admire your dedication, it's a rather simple repair that shouldn't be all that hard.

Dumb question, but how do you test components that are part of a circuit? I know I can use a multimeter to measure resistance when the resistor is out of a circuit, but will it give the same reading when a resistor is in a circuit? Same with testing capacitors, diodes and other components. Or are you saying the schematic should contain test points and marked voltages?

 

[Nicholas Lindan]

Most likely the only failed components will be the electrolytic capacitors. Looks like they used pretty cheap ones. The other likely failure would be the Q4008 triacs. Also replace anything that looks burned or where the board has become discolored.

I would start by limiting the replacement to these parts and see if the boards come to life.

80's vintage semiconductors should be perfectly good. Replacing IC's, unless you have a something like a Pace desoldering station, is not a good idea as there is a very high probability of damaging the circuit board, especially true of double sided boards with plated through holes. If you must replace them then cut the legs from the body at the top of the board, heat the pin from the bottom and pull each pin from the top. Use a solder-sucker or braid to clean the holes out.

There is an engineering adage: "Fix anything long enough and you will really break it."

 

[Nicholas Lindan]

Dumb question, but how do you test components that are part of a circuit? I know I can use a multimeter to measure resistance when the resistor is out of a circuit, but will it give the same reading when a resistor is in a circuit? ....

http://en.wikipedia.org/wiki/Bed_of_nails_tester

 

[epatsellis]

Pretty much as nicholas stated, though I have a logic analyzer and 2 oscilloscopes on my bench as well. Well, lots more, but you get the idea. Desoldering double sided plated through holes without the proper tools, well you'd better off just dousing it in lighter fluid and save yourself the frustration.

If it were on my bench, I'd isolate the fault, figure out which part of the circuit is responsible for that function and do a little hunting, after I replaced the electrolytics, as Nicholas suggested, you will likely find it's a simple failure of a $.10 component, at best.

The likelyhood of the bulk foil caps failing is about zero, maybe a little less.

 

[Bob-D659]

I'd give the connector pins a very close inspection, even to the point of resoldering every one of them. One of our cars had a digital instrument cluster, after a few years they started getting errors. Replace it or pay $1,100 for a rebuild. Well it turns out the design spec was so cheap the harness connector shells had no physical attachment to the circuit board, just the solder joints on the pins. Now how much stress does a wiring harness put on a circuit board in a car? Esp when you have to pull and twist on the cluster to wiggle your hand in behind it to remove the connectors to replace a burnt bulb. Resoldering the connectors cured all the intermittent errors in the cluster.

 

[epatsellis]

Bob, you may want to see if you can reinforce the pins by damming and flowing epoxy to add a structual element to the connector (the one that was cost engineered out...)

 

[Nicholas Lindan]

I'd give the connector pins a very close inspection

That problem isn't so bad with PTH 2-sided boards, but is a big problem with single sided boards.

If a single sided board fails then 99% of the time the problem is a cracked trace or solder joint. Often the connector pin or the lead on a heavy component lifts the pad and the foil cracks where the trace meets the pad. The other problem is that there is not enough solder to form a good fillet between the pin and the pad and an annular crack forms where the solder bridges from the pin to the pad.

Examine the pads around the transformer pins for solder cracks, broken traces and lifted pads. The transformer pins should be clinched (bent over tightly) as there is no hardware securing the transformers to the board.

 

[Bob-D659]

Bob, you may want to see if you can reinforce the pins by damming and flowing epoxy to add a structual element to the connector (the one that was cost engineered out...)

No need, the tranny is 99.99% worn out and the car just had it's 21st birthday.

 

[ic-racer]

Hey guys thanks for the posts. Realize that in the first post some of the 'problems' with the units are intermittent. For example, the one that blows bulbs is adjusted right to spec at 80.0 v and the few other voltages on the schematic are in spec. Bob-D659, on that board, BTW I did find a cold solder joint on the connector leading to the lamp. I also put in a fresh TRIAC, and saw that the old one was not making good contact with the heat sink. I used that board last night and no blown bulbs so that board may have had its life-span extended.

The schematic does not give any useful voltages or document appropriate signal behavior. In terms of the service manual, the board is treated as a 'replaceable item.' Another issue is that access to the board is very limited once it is in place in the head. Extended jumpers and 'breakout' boards would need to be constructed to do any major diagnostics to this board. I did look into turning my 'spare parts' head into a test dummy just to test this board as I agree with you all the it is probably some 5 cent part that is amiss, however, constructing a 'test head' would be quite a project in itself.

As I stand now the 'spare' board is already stripped and IC sockets are in. So once the parts arrive I just need to plug in the ICs and add the few resistors and caps.

 

[ic-racer]

Just for interest, here is the schematic for the board.

 

[Mike Wilde]

I second the comment about the electrolytic caps. I have rebuilt three Vivitar timers and process controller power supplies, circa early 80's The caps dry out and go flaky. The other area of concern to look at are zener diodes, and thier pull down/pull up resistor. If the resistor at OEM time is too small, it will burn out and all sorts of wierd voltages then appear. I have replaced them with voltage regulators.

 

[ic-racer]

Another thing that may not be evident (unless you own a D5500) is that this power supply board lives up in the lamphouse. In fact the thermal overload thermistor is on this board.

That big orange electrolytic capacitor has a date code of 1983 on it and this unit came from a busy lab. So, I suspect every component on the board has had a rough life.

I was looking over the components I removed and I did find a fried 27 ohm resistor leading to the power pin of the IC U14, which contains a bank of general purpose transistors. It still reads 27 ohms and its not likely the cause of the light not working, but just an example of how every component on the board has likely been stressed.

 

[frotog]

Reconditioning a 5500? Isn't that a little bit like doing a ground up resto on an AMC pacer?

 

[BobNewYork]

Reconditioning a 5500? Isn't that a little bit like doing a ground up resto on an AMC pacer?

To us D5500 aficionados them' fightin' words

Bob H

 

[ic-racer]

Reconditioning a 5500? Isn't that a little bit like doing a ground up resto on an AMC pacer?

I do not know (the last car I restored was a 1979 Ferrari) but some one must know the answer

 

[Bob-D659]

Factory air I see. Looks like you're missing the optional chrome roof rack tho.

 

[frotog]

To us D5500 aficionados them' fightin' words

Bob H

I guess that depends on how you regard the Pacer. Me, I always thought they were strange but cool cars. Definitely not a great performer. Sort of in line with everything I've experienced with the d5500 head and controller. It seemed like every lab I've worked at has had a few of these heads, inoperable, half-gutted, lying in boxes. Closed-loop is compelling but only if it works. Aren't they notorious for shutter failures? Sporadic controllers? Bad electronics? Is the d5500 really an improvement of the time-honored chomega ii?

 

[ic-racer]

It seemed like every lab I've worked at has had a few of these heads, inoperable, half-gutted, lying in boxes. Closed-loop is compelling but only if it works. Aren't they notorious for shutter failures? Sporadic controllers? Bad electronics? Is the d5500 really an improvement of the time-honored chomega ii?

I have thought about this and this is my conclusion. When an 'Enlarger Company' makes a new enlarger, the R&D pretty much ends once its in production and they sell the same thing for years and years. This is actually a good thing because it means to be successful the have to get everything right the first time and also the long production run means parts accessories will be available (think D5, Chromega II and Besseler 45 etc.). However, once an 'Enlarger Company" gets in to closed-loop electronics you would have to keep up the R&D to keep up with the electronics industry. For example the D5500 has a port to connect to a Commodore 64. That system was still in place when they were selling new D5500s in the late 90s. From what I have seen, the electronics in those D5500s sold in the late 1990s were assembled from components they bought back in 1982 when it first went into production. I think that Enlarger Company philosophy of selling the exact same thing year-to-year without electronic upgrades lead to the problems that you have mentioned.

There actually are some good things to the 'outdated' electrical design. The circuits are on conventional PC boards with 'standard' DIP integrated circuits. Just about any contemporary electronic device today has a circuit board made with SMD components and these are nearly impossible to repair or modify. Also, the D5500s primitive origin means its circuit is less complex and easier to understand and troubleshoot.

If I were to use an automobile analogy, though, I would have compared the D5500 to something like the Bricklin or the DeLorean. I think you will agree that the D5500 was a pretty modern design for 1982.

 

[ic-racer]

I am expecting the new components in the mail today. Here is the circuit board ready for the new stuff. That round component with ten legs is still in place. That is the $13 TRUE RMS converter and I won't take it out until I actually have the new one in my hand.

 

[BobNewYork]

I guess that depends on how you regard the Pacer. Me, I always thought they were strange but cool cars. Definitely not a great performer. Sort of in line with everything I've experienced with the d5500 head and controller. It seemed like every lab I've worked at has had a few of these heads, inoperable, half-gutted, lying in boxes. Closed-loop is compelling but only if it works. Aren't they notorious for shutter failures? Sporadic controllers? Bad electronics? Is the d5500 really an improvement of the time-honored chomega ii?

I liked the closed loop - it always worked for me without a problem. All it means is that the mixing box automatically compensates for the gradual changes in your lamp and filters and once in a while you zero out and re-calibrate. I have to say it worked very well for me from '86 when I bought it until the fire department hosed it down last year! And I always thought the D5500 was reliable

Bob H