68kMLA Classic Interface

This is a version of the 68kMLA forums for viewing on your favorite old mac. Visitors on modern platforms may prefer the main site.

Click here to select a new forum.
Mac SE (or SE/30) Help Needed after extensive CRT Horizontal Drive Testing and Repair
Posted by: powellb on 2026-09-26 13:20:04 This is a lengthy post. TL;DR The horizontal drive on my SE analog board has failed. Components all check out, but increasing inductance improves it while distorting the screen. I previously posted around 6 months that my SE horizontal drive failed so the screen was stretched horizontally (only 50% of display shown as in the image below). In addition, the brightness was greatly diminished. I left it for awhile, but I have been spending many hours on it over the past month. Let me go through what I have done because—at this point—I no longer have any idea how to proceed. I am hoping an analog expert can chime in. I will also note that I am using a Babyface PSU, which does not provide "12V sweep" power, but standard sweep power (as sweep power is intended for horizontal drive). I don't believe this is the issue because the sweep power was independent of the other 12V so hard drives, etc. wouldn't cause the CRT to change. I don't have anything installed right: my SE/30 motherboard and CRT. I am open to suggestions. Tests and State of Board Let me first state that I had previously (18 months prior to this issue happening) replaced capacitors, a couple of resistors, fan, and reflowed the board. Since the issue, I have been examining the attached schematic (sorry for the poor quality, it was the best I could find). The horizontal drive is also tied to brightness; hence, losing brightness with loss of proper horizontal drive. I think it turns out energy isn't flowing properly through the horizontal drive rather than brightness, which I will get to. CRT: I have two Samsung 9" CRTs. Both exhibit this failure, so I am confident it is not on the tube side. I tested the 74LS38 that controls the horizontal sweep. Works fine, on the scope, it is driving at 22.25kHz. The 74LS38 drives the main transistor, BU406 to drive the flyback at 22.25kHz. I replaced BU406 with new part. The scope shows that flyback (T1, pin1) is driven by BU406 correctly with the right frequency. I removed the flyback (T1) from the board and did a ring test using my scope. There were 6-7 decaying cycles after each pulse, so it doesn't seem that the transformer has any shorted loops. Checking the winding resistances, etc. showed no issues. After the flyback, I moved to R18 and R19, both removed from the board showed the proper resistances (100 and 220 ohms, respectively). To be safe, I replaced with new resistors. R19 works in parallel with the L3 inductor. I can't seem to find official specifications for inductance for these custom Apple inductors. As near as I can tell, L3 is specced somewhere between 8-15 µH. I pulled mine from the board, and it tested around 9 µH. CAVEAT: as we will see, I think these are non-linear inductors, so the measurement (from a standard LCR meter) is not accurate since it is only at a single current. L3 seems fine. C12 and C15 in the path had previously been replaced, but I retested, all perfect. Tested the forward drop, zero reverse voltages on the diodes, no issues. We come to L2, the variable inductor to adjust horizontal width. I removed it from the board, and it tested as a range of 37-44 µH depending on the slug position. This does not correspond well with ranges people had discussed, which seems that the slug should move it from ~15-50 µH. The odd thing is that with L2 being skewed high, I would expect my horizontal drive to be too narrow. The overriding test though is that if I insert a stainless steel rod into L2 (to increase inductance), then the horizontal drive fell nearly into spec. So, L2 is my main concern, and it is a common fault. All solved? Measuring inductance of L2 without and with the rod in the core shows a difference of ~20 µH. So, it seems that I need extra inductance in the drive. With the failed horizontal drive, I also tested the voltages at pins 1 and 7 of connector P2 (neckboard connector) to check the brightness. The specs are 38V and range of -25 to -70V for pins 1 and 7. I measured 63V and -167V while the horizontal drive is not in spec. When the rod is in L2, these values fall back into spec. So, with inductance too low, it seems that the horizontal drive isn't using the energy produced by the flyback and it has to go somewhere. Repairs I have Attempted Since L2 (or, as I will come to, the combined L2 + L3) inductance is not correct, I tried to source a replacement. The closest I was able to get was a 128k analog board for a few bucks. I tested its L2, which showed a range of 16-48µH, so that is in spec. I swapped it into the board, and: No Change. Horizontal drive was not improved. This makes sense in that the original L2 was stuck around 40µH, and the replacement L2 max was around the same. I then checked L3. The 128k L3 part showed 17µH as compared to 9µH; however, as these are nonlinear, I don't really trust the measurement. Regardless, I swapped the 128k L3, and again no change. Since the only thing I know from the measurements is that an additional 20µH makes a difference, I installed a new solid-state 25µH linear inductor between C15 and L2. This had no effect either; however, I also realized that the original iron-core inductors have resistance with them. So, I installed a 1ohm ceramic/cement resistor (to handle several amps) between the new inductor and L2. THIS WORKS! So, now I have a working display (albeit with a horrible conglomaration of two new, large components soldered to the backside of the board. All done? Not quite. The issue is that I've added linear inductance, so while the width is the correct size, the information is different. It seems the sweep accelerates across the screen, so the left side has wide pixels and right side has very narrow pixels, and there is horrible ringing on the left side where the gun returns to start a new line. The final image below somewhat shows this effect. It is very pronounced as the cursor is large on the left and more like a vertical line on the right. Questions Hopefully someone can help me with the following: All of my original inductors seem close-to/within spec, so why do I need to add an additional 25µH with resistance to get it to spec? If I do have to correct overall inductance, how do fix the linearity issue? I did try a 27µH iron core inductor that was on the 128k board to replace the 25µH/1ohm resistor, but it compressed the entire horizontal drive to about 40% of the CRT area, so it overcompensated. It also had horrible ringing and distortion. The first question is the most important, because I feel like there is a fundamental issue somewhere, and I am throwing inductance at it to mask it. I appreciate any thoughts/comments.
1