Phoenix Project: HK Citation 7.0 A/V Controller

[sleddriver]

Good old Made in the U.S.A. audio gear. I bought this piece used many years ago at a substantial discount from its new ($3450) retail price. It was manufactured at the now closed Madrigal plant in Connecticut where they also produced Proceed.

I used it daily for nearly 15 years before one day it failed to light. The problem turned out to be an overheated & burned-out main power supply transformer. With Madrigal long out of business, I thought the Citation 7 was too. A search turned up no replacement, so I thought of assembling a new power supply in another box. Then while searching for transformer manufactures in the USA, I discovered the company that made the transformer for Madrigal was still in business. After talking back & forth and many emails, I placed my order. After a long wait of 4 months, I had a new one in hand.

Now, back to the board photos..further investigation revealed long-term heat damage to the main PC board. This is best revealed here in backlight, which is very effective for finding board hot-spots. Notice the yellowing of the blue fiberglass board on the left hand side. The three main electrolytic capacitor terminals can be seen. The extreme brown area below them is where the main power diodes were mounted...right close to the board (big mistake). They need to be up a bit to provide better cooling and prevent this kind of heat-related damage.

Zooming in here, you can also see signs of over-heating on one of the regulators mounted to the heat sink. It's the three vertical pins to the right of the PS main caps.

Here's the view from the other (component) side. I've removed the power diodes, thus the empty holes. They'll be replaced, along with the smaller units to the left and the the main PS caps in the background. Surely they must be 'cooked' after years of exposure. The PS transformer mounts below.

Note the very large heat-sink for the main regulators.Too bad the diodes weren't heat-sinked.

Here's a section of the video switching board showing heat damage from a regulator. It's easy to tell which one of the three is drawing the most Watts. I'll be replacing each of the regulators, their associated caps and resistors.

Here's the same (dusty) board front-lit.

Here's the main board front-lit. The layout makes troubleshooting much easier. Madrigal did a nice job of labeling various sections of the board and the schematics were still available from Harmon Int'l.

Here's some front-panel photos.

I'm looking forward to getting her lit again and enjoying surround sound, now that the replacement transformer has been received.

Stay tuned for updates!

 

[dparm]

Whoa, that thing is awesome. I'll be watching this. Makes me feel so poor with my Marantz...

 

[sleddriver]

Update:
Heated up the iron this evening and pulled a few electrolytic capacitors for a value check. First up where the three 4700uf/50V main caps. I'm not able to measure these with the Wavetek I use directly as it only measures up to 2000uf. However, given they lived a very hot life next to the power diodes, they'll be replaced with new caps.

Next, I pulled several 220uf/25V caps: They measured 146, 134, 95, 82, 22 & 5uf! 10% low would be 180uf. 20% would be 176uf. The highest one is 34% low. The rest are even worse. The Citation has 20 of these. I'll bet they're all bad.

Next I tested some 1000uf/16V units: 400, 457, 455, 470, 554, 590. Way low. Again, 10% tolerance would be 900uf. 20% would be 800uf. These are more than 50% low! There's 13 of these to be replaced.

On the video switching board, I found some small 10uf/50V caps that serve as the main on-board filters for the +- 5V supplies. All three were a deep orange color. They measured 0uf, 0.06uf, and 0uf. Tango Uniform. They weren't filtering anything. This explains the sqiggles I would see in the background, particularly when the menu overlay was called up.

On the final regulator, it looks someone had replaced the OEM caps with a pair of Panasonic HFS 100uf/63V 105°C units. These are very nice and still holding up well despite their age: 94uf & 102uf. Excellent. Not sure when these were added but it must have been at least 16 yrs. ago before my purchase. Panasonic makes some great low ESR, high temp EC's (electrolytic capacitors) that age very well.

Not much use in showing photos at this stage because it's just meter reading. When I get CLIO up and running on an older machine, I'll do some frequency sweeps of these old caps vs. new.

This is SOP for reworking old & older electronic equipment. As EC's age, the paste inside dries out, their value drops, they become more inductive and begin to allow more ripple onto the DC power rails. Plus, they begin to run hot. So it's a good idea to pull some and test them. Depending on the quality, they may be just a bit low. But even this expensive piece of USA audio gear only used marginal EC's. They fail miserably compared to the two Panasonics.

Another common item to replace are any charbroiled resistors (usually an indication that the wheels came off somewhere) and any small incandescent lamps that have burned out. The Citation doesn't have any of the later, and was one of the first I think to use blue LED's which were quite expensive in the early 90's. That's all for now.

 

[sleddriver]

Update II:
So far I've removed 79 EC's, cleaned each hole of old solder, installed new caps and soldered them in place. Since each cap has two leads, that's 158 separate connections multiplied by 3 operations/connection. It gets tedious to say the least....

While installing new power diodes, I noticed significant damage to the copper traces due to overheating. Thus I'll have to suppliment them with individual wires. You can see this area in the photos.

I'd post some pictures, but not much to show except for a pile of old parts.

I did fire up a very old confuser that I dedicated to running CLIO, which is an audio analyzer with LCR meter, FFT, CSD and waterfall capabilities. I thought it might measure the 4700uf EC's. No joy.

Still it was a hoot to see this old thing come back to life. It's running W98se. I had to remove the cover, wiggle some wires, unplug & replug, discovered again it only has ONE round-style serial port. I can't even remember what you call them! I searched around for a 2 to 1 adapter, but no joy either. So I lit her off with only an (old) mouse attached and after several falst starts, I found myself staring at the W98 desktop...on an old crt monitor to boot!

I recallibrated the A/D, D/A card and measured some smaller caps with it. They were in agreement with the DVM. I also did a few Z vs. freq. sweeps. Works fine.

One thing that has helped immensly was to take the .pdf schematics down to a copy shop with a large format printer and have the main board schematic & parts layout expanded to a 24" x 36" plot. Much easier to read that way. I also plotted several of the smaller sections onto 11 x 17 sheets. Sure beats squinting & using a magnifier.

The next audio project is to re-lamp and replace EC's on an old classic Pioneer SX737 stereo receiver I've owned since 1978. It still works fine, but most of the front panel lamps are Toes UP so I have to use a flashlight to see the tuner scale. As most boards are connected with wire wraps, I'll have to dig out my wrap/unwrap tool.

More later.....

 

[sleddriver]

Update III:
Installed the new PT and temporarily bolted it to the PCB to remove any torque from the pins while the board is vertical. Due to the poor condition of the pads, I used some small push-in terminals to take up space between the PT pins and the pads and give me a place to attach some wiring, which was directly wired to the pins of other devices. Once all wiring was in place, it was quickly soldered with a hot, well-tinned iron.

 

[sleddriver]

Update IV:
Citation 7 Back On-Line:

Current Readings (with video board plugged in; All temps in °F):

Line Voltage: 128VAC(!)
VARIAC output Voltage: 110VAC
AC current draw: 0.5A
AC Power drawn: 47W
Main power transformer case temp: 120°F
Main heatsink temp: 105°F

Video Board:
-5V pre-regular heat-sink temp: 106°
-5V regulator HS temp: 110°
+5V regulator HS temp: 110°
V drop across -5V pre-regulator:_____
V drop across +5V regulator: 12V

What's interesting about the C7 is that the front-panel power switch doesn't turn it off. The power supply is still fully active and as shown above, runs hot. Either there is a faulty component pulling excessive current from it or HK decided to Class A bias all the audio & video drivers and buffers. Further by leaving it "all on, active & hot" it's ready to go whenever you are.

Problem is it is dissipating quite a bit of heat 24/7. There is a true power switch in the rear. I found in the past that when used, the programmable settings are also erased. Which means setting it up from scratch all over again.

The solution is to keep the memory alive but turn off the rest. That will take some modification, but in the long run will prolong component life and eliminate the 24/7 production of heat. While this may be considered a 'feature' up North in colder climates, here in Texas we already have an abundance of it and have to pay monthly to remove it.

Next step is to take DC current readings on each stage and check each IC for hot ones.

 

[HTSS_TR]

I wish I understand the degree of difficulty of what you're doing !

 

[OVERKILL]

This is an involved project! Kudos to you for doing this though

 

[sleddriver]

^^Yes it is. And now it just became even more involved when I discovered hum/buzz on the audio outputs and no video pass-through nor internal character generation! Rats. I'm currently guessing that when it went Toes Up, it took some silicon with it. The regulators are functioning at their designated output voltages, but I haven't checked for ripple yet with the o'scope. Could be that some/all of the regulators went TU when they overheated. I actually meant to replace all of them originally as they're not very expensive.

Yesterday while testing, I noticed the main PT wasn't nearly as hot as the day before. There is a raw +17 volt line to be regulated to +- 15V for the high current buffers and the rest runs off of multiple 5 volt 3-terminal regulators. With only 5V though, something is pulling a lot of current to generate that much heat. Could be a transistor is shorted somewhere. The scorch marks on the video board mentioned above is a good clue.

The C7 uses a number of both 4051 & 4053 analog mux/demux's to route signals throughout both boards. These are both CMOS, though there are a number of them so current draws add up.

It would be very helpful to have the use of a FLIR camera to see which board areas are dissipating the most heat. Baring that, even an inexpsensive non-contact IR thermometer would be helpful. Perhaps it's time I buy one.

It's time to warm up the oscilloscope and have a look. Thanks for reading.

 

[sleddriver]

Update V:
Using the oscilloscope to view the AC ripple on the output of the various regulators revealed that one in particular was allowing far too much ripple through. I replaced it and the one next to it, also a +5 regulator. Another check on both of these revealed negligable AC output. Just nice smooth +5VDC.

I also checked the second discrete diode bridge and found some surprises. More than a few were leaking when reverse biased. They also showed an elevated resistance when forward-biased. These were very small 1N4003 diodes with a current rating of only 1A, undersized IMO. They were soldered right down on the board where their heat dissipation turned the board brown as shown earlier. I elected to replace them with larger 3A versions and set them up away from the board to allow better heat dissipation.

However I had to stagger their heights so they would all fit in the confined space, much like you have to do with several helium party balloons: Some are on shorter and longer strings to keep them from all bunching up and spreading out.

Replacing this bridge opened up another can or worms because of pad damage to the overheated board...they just came off. Thus more custom discrete wiring to make up for pad & trace damage on both top AND bottom sides of the board. I rechecked this multiple times to be sure I didn't make a left hand/right hand mistake as the view is different from the top vs. bottom.

After reinstalling the main board (with the video board and front panel display/input switcher disconnected), I fired her up with only 50VAC which only provided enough power to bring the 5V regulators to about 3.5V. This allowed me to check polarities and for hot regulators before slowly increasing the voltage. Fortunately I made no mistakes with no TATO incidents.

Another round of AC, DC voltage measurements revealed all was well and stable. The o'scope initially showed some ripple on the outputs of the +15 and -15 regulators. Once they heated up though, it disappeared into the noise floor. Unfortunately I'll have to order these as none are available locally anymore. The remaining 5V regulators all showed no AC ripple on their respective outputs.

I made the following power & temp measurements (in °F):

AC Line V: 123Vac
VARIAC output to C7 V: 110VAC
AC current draw: 300mA
AC power drawn: 27W
Power factor: 0.8
Main PT case temp: 100°
Main Heat-sink temp: 100°
A/D converter regulator heat-sink temp: 109°

These will no doubt rise once the video board is reconnected. Right now though I'm troubleshooting one board at a time. Getting the PS stable with no ripple is key.

I also checked all audio I/O jacks and measured no hum nor DC. Previously there had been noticeable hum on the front L&R outputs. This is a great sign!

So far so good...I powered down, installed & connected the front panel/controller and powered up. If you'll refer back to Update IV, you'll notice 4 green LED's on the right side of the display panel. These are L & R input level indicators and the top one lit refered to 0dB. The was the hum/buzz I referred to earlier. Now it's gone and these LED's are extinguished! The o'scope confirms this..all outputs have no DC and the AC is in the noise floor.

So what this means is that I should have replaced the additional bridge and all V regulators while I did the EC's. I had originally planned on it. Just found myself in a deep well after hours spent comparing EC's and preparing an order.

Regarding the video board...my small breadboard supply wasn't capable of supplying enough current for it. My other supply isn't capable of simultaneous double-ended operation: I thought it would! So, I need a better bench supply.

Now that the main power supply is stable and noise-free and the audio section output is clean, I can proceed to troubleshoot the videoboard.

Stay tuned and thanks for reading...

 

[sleddriver]

Update VI:
Spent some time today troubleshooting the video board. There are three separate regulators for this board, supplied with power from the main PS. Both supply lines contain 100uH inductors to damp noise that might be picked up from the 1x16 ribbon cable between the main PCB and the VB.

Close inspection revealed both of these had overheated a little as I noticed a slight brown tinge to the vinyl covering. Further, one of the inductors had a small crack in the outer vinyl covering at one end, exposing the tiny, fine wire wrapped around the core.

Above the 100Ohm resistor, you can see where the brown vinyl jacket has begun to split on the left end of the 100uH inductor. Neither of these inductors look factory installed as the length isn't quite correct. There was enough heat dissipated over time here to brown the board below slightly. I'll be replacing both with units designed for higher power dissipation and levitating them a bit above the board.

Beneath there was more pad/trace damage which had been repaired with a single wire in parallel. During testing, this inductor went Toes Up and tripped something else, causing the front fluro-display to latch up. Cycling the power didn't help, but removing the two ribbon cables between the boards did.

I soldered a short across the open inductor and wired the VB directly to the main PS. All the VB regulators were working fine, but I noticed the +5Reg heat sink heating rapidly. It topped out at 150°F! It was still operating though as they can run up to 150°C...emphasis on °C. But I didn't want to go there so I pulled the plug.

The reason for this is because the resistance of the inductor was replaced by a dead short, thus applying +20VDC to the reg. input while it continued to regulate its output to +5V. Therefore +15V was being dropped across it instead of around +11 as usual. Result? Lots of heat dissipated. Until a replacement arrives, I sub. in a resistor.

Not sure what's going on with the VB connected. Before this incident, I measured all video outputs with the scope and there was no noise, which is good.

Due to the holiday, I'll have to wait until Tues. for additional parts. I'm gaining on it though! Stay tuned....

 

[sleddriver]

Update VII:
Replaced both of the 100uH inductors shown above with new units. After wading through the manufacturers data sheet, comparing size vs. board room vs. power dissipation, I couldn't meet all criteria. Max DC current with a slightly bigger size is 151mA. As fate would have it, this still isn't big enough. More later...

Also replaced the last of the JRC 7905 plastic case -5V video board regulators with a TI model in a cer-dip case (better heat transfer to the heat sink).

I fired up the PS and measured -19V and +9.3V(unregulated) that feeds the VB regulators, which at this time remains disconnected. The schematic indicates these should be closer to -15V and +5V respectively. I'm still using the VARIAC to keep the VAC into the PS at 110VAC. At the wall I measured 123.5VAC.

As if these voltages aren't already high enough, they'll be even higher once the unit is plugged into the wall. This tells me the power transformer is out of spec with both secondaries having too many turns. Nice... I'll have to bring this up with the transformer manufacturer. This is a clue as to why the original transformer and the C7 in general ran hot. Heat dissipated goes up with the sqr of the current. Things that run too hot for too long tend to go Toes Up quicker than necessary. This explains why the early photos show the board 'browned' in spots and why the inductor I replaced looked overheated.

With the VB reconnected and power recycled, the front panel lit all of her mode-indicator lights, and with only a "_" on the fluro-display. I'd seen this before. Pulling the VB data connector before power cycling would always work. Not quite sure why though as up until that time, I haven't paid much attention to the front display PCB and circuitry.

I did go back and recheck the three 10-pin connectors between the main PCB and the front panel PCB. Under magnification, I found a few that were making poor contact with their ribbon cables. After I tweaked each, plugged in the other data cable to the VB and and applied power, the front display came back to life. Good!

Next, I checked the operation of all front panel controls and all buttons on the remote control. All were OK. Another good sign.

Then I noticed that one of my two new ceramic 100uH inductors was beginning to turn brown! Uh oh....it doesn't feel hot though. Next I noticed that a seam had opened up on the side and a small drop of _____ appeared! You can clearly see the color difference here:

Next I checked the temps of the heat sinks:
Main: 106°F
Power transformer: 118°
+5 VB regulator: 103° (the above inductor connects to Vin).
-5 VB pre-regulator: 126°
-5 VB regulator: 113°

Note that these temps would all be higher if the C7 was plugged into the wall, being fed with 123.5VAC instead of 110 because the greater the voltage drop across the regulator, the more heat it dissipates: P = V * I or I^^2 * R or V^^2/R.

I might remove the browning inductor and replace it with a metal-oxide power resistor of a few Ohms. Matter of fact, I may have to do this on the + VB rail as well to reduce the voltage across the regulators to keep their heat dissipation in check.

I've left it on now for 1.25hrs and the front display has locked up again: It won't respond to any front panel nor remote control contacts. I checked it at an hour and all was well. Perhaps something is slowly overheating...might be a leaky transistor or capacitor in the front panel. Upon cycling the power, all is well.

Next step is to hook it up to a TV and see if the on-screen menus appear. No joy here. I see what looks like oscillation: two horizontal bars that are a bit fuzzy. Just touching the main composite video out causes the C7 to reinitiallize or light up all of her front panel lights.

So far, my diagnosis and repair has focused on the various power supply main rails and replacing all the main & video board EC's, as well as a few miscellaneous parts, and repairing several board traces. There is still the audio, video and digital switching sections to diagnose.

More later...

 

[sleddriver]

Update VIII:
It's occured to me that I need to post a pic of the schematic here. The VB PS is shown at the bottom. Note the location of the two 100uH inductors and the zener lifting the 1st neg. reg. above gnd. The left edge shows the two connectors that carry power & data between the main & VB. Above & to their right are the three PAL's mentioned further down. The STROBE IC unfortunately overheated and went Toes Up. Finally on the right side are a few of the video output drivers.

I found out that on the video board circuits in question, attempting to measure resistances on resistors, capacitors and some transistors results in false values. Thus I heated the iron to 'lift-one-leg' for the bi-terminals and completely remove the tri-terminals. I thought I might have been on to something, but once removed all of these pieces measured OK.

I pulled the browned 100uH inductor shown previously and measured it's resistance at 0.7Ω. Originally it was 4.1, so it's TU. I temporairly replaced it with a 5Ω 5W resistor with a leg-high to connect a mA meter.

Upon pwr on, the front display still showed the status lights all lit and a solitary "_" on the fluro-display. No joy. Current through the branch from the main PS to the on-board +5V reg. was 200mA. No wonder the inductor only had a short life...it was rated for 150mA as was the original I replaced.

Likewise there is a 100uH inductor feeding the two -5V regulators. It's still showing no signs of distress. I haven't measure the current flow on this side yet.

So it seems something is amiss, drawing too much current through this part, which is before the +5V reg. I may have found a solution though.....

I started checking an area of the VB which contains a number of cheap trim pots. Often when something begins heading for the ditch, too much current will be pulled through one of these wipers and take it along for the ride. When that happens, sometimes something else down stream will also be taken out. Since I wasn't getting anything out of the character generator IC, I looked farther upstream to several of these. This is where I measured in situ mentioned above and did the "leg-lifting"....and struck out.

After reinstalling the VB, hooking up the ammeter and powering up, I checked voltages again and used the o'scope to verify none of the regulators was oscillating and the ±5V rails were clean.

While probing around, I noticed a small burn mark had appeared in the middle of a DIP IC labeled "STROBE". Very odd. There are three of these IC's total: CLOCK and CONTROL. Since they weren't ID'd on the schematic I removed another label to find PALCE 16V8H-15 EE CMOS Universal Programmable Array. These have code burned into them at the factory. I quickly turned off the power but think I'm too late. Appears to be an internal short. Not at all sure Harmon still stocks this part.

The fluro-display had been acting up. However after removing this IC and powering up, the display looked fine and all the buttons worked. It's been on now for an hour and still hasn't latched up.

I used the o'scope to view some of the signals in the digital domain, beginning with the heartbeat: CLK. Plenty of ringing. The DD likes a very clean CLK with sharp edges. This can cause false triggering, corrupt data, framing errors, and maybe overheating.

V: 1V/div; H: 0.5uS/div
Before removing the burned IC, the ringing on top was as bad as on bottom.

Obviously this part was drawing too much power due to an internal short. With it gone, the VB is unable to switch any video outputs. However, the rest of the C7 appears to be working fine. Further, it may even be possible to just unplug the VB and still use the audio section.

From the looks of it, I won't be enjoying any surround sound any time soon. No doubt I was hoping for the best by replacing the PT and some EC's, and having all return to normal. Unfortunately it appears that some other circuitry might have been taken out and damaged as the wheels began to come off and she headed for the ditch.

I'll see if Harmon has any parts lying around or I might get lucky on ebay.

 

[sleddriver]

Update IX:
After a bit of a search, I discovered something better: The whole video board! This will give me one to compare mine to and also provide some additional parts for spares. I expect shipment soon and look forward to its arrival.

 

[sleddriver]

Update X: NEW VB INSTALLED AND FUNCTIONING! Yee-Haw!

Finally had some time this afternoon to install the new VB and stretch some video cables across the room to verify proper operation. YES! Old video distorsion, squiggly lines in both the video pass-through signal and from the character generator ckt is gone. Verified proper operation in all 8 inputs. All were clean.

Next browsed through a few menus, made some changes, powered off, then re-lit to verify they were stored. Yes! Was going to post the boot-up status message, but it only appears for 2sec.

It's been on now for 40min. and no issues. Here are some vital signs:
VARIAC output: 114VAC
AC current drawn: .42A
AC power drawn: 41W
PS Trnsfrmr case temp: 120°F
Main HS temp: 121°F

Video Board:
-5V pre-regulator HS temp: 124°
-5V regulator HS temp: 113°
+5V regulator HS temp: 114°
+rail inductor current: 214mA; Pwr diss: 146mW
-rail inductor current: 184mA; Pwr diss: 109mW

The power dissipated through the +rail 100uH inductor is at its max limit. This explains why this part was replaced on the older board and the brown PCB discoloration due to heat. Further, it's right down close to the board with no airspace. Notice the -rail inductor power dissipation is ~25% lower and below its max. pwr diss. limit. Both are warm to the touch, but not hot.

Another issue related to the older VB's failure are the 85°C filter caps located too close to the regulators HS's. Unfortunately I can't bend them away because the axis of the caps leads is perpendicular to the plane of the 3T regulators. This is the reason the others turned from yellow to 'burnt' orange, dried out, their capacitance dropped to near zero and allowed noise and ripple through into the regulated VB power supply rails. In addition, some 3T regs are not stable without an EC on their output. I think this caused the unstable character generator display I remembered. I'll need to experiment with possibly relocating these EC's nearby to get them away from the heat and prolong their life.

I'll need to record critical voltage measurements for future reference and upgrade some of these parts. I have yet to test the audio board, surround sound, audio input switching, etc. so it'll need to be done before reinstalling the Citation into the equipment rack.

This is all very good news though as it's been quite a climb to get to this point with the VB stable.

Stay tuned...

 

[sleddriver]

Update XI:
Having come this far, I've come upon a stumbling block regarding the Citations design and I've been considering possible options to work around it.

I discovered some previous notes of my own that I'd completely forgotten about regarding troubles with the video board. I had replaced two of the EC's around the -5V pre-reg and both of the 100uH inductors mentioned earlier because all had over-heated. The former caused distorsion in video signals, according to my notes. When the C7 went off-line a few years back, the same problem was occurring due to the other EC's around the -5 & +5 regulators going Toes Up.

Further, the reason for all the brown PCB areas shown in the beginning photos is because when the front panel or remote power is turned off, it's not really off. Not even close to off. There is no 'low consumption' mode...not even close.

I verified this by touching the front panel 'power' button, then noting any changes in supply voltages and HS temps. There wasn't any change. This means it's running at the recorded temps 24/7, whether it's ON or OFF. At 120°, the PS transformer itself is so hot you can't touch it for longer than a few seconds.

The -5V pre-reg is even higher at 124° and is radiating heat 24/7 directly to the EC's located only 1/16" (worst case) or 3/8" (best) away. The OEM caps were only rated at 85°C or 185°F. They should have been rated at 100°C (212°F) and preferably located further from the HS.

The two 100uH inductors are undersized and located right on the board. I measured about 210mA of current through the one that cracked. The one on the second board looks like a 1/4W resistor. It might handle 140mA max. The replacement needs to be at least rated for 300mA for sufficient margin.

All of the 'scorched' areas seen in the photos are due to the PS being ON, even when the unit is OFF, and are located underneath significant heat dissipating components.

There is a master, physical, analog main-power-switch on the rear panel which really is OFF. So you might think "OK...so use that." The problem here is that when you do use that, all the data you entered in, speaker distances, types, surround sound details, labels of inputs, custom modes you've set up.....disappear.

Given the Citations 7-inputs, this can be a lot of info to re-enter just to use it. In my case I mostly use three sources all of the time.

Options are to modify the PS distribution system so that only the memory is kept alive 24/7. I'm not sure this is possible given the design, but it's worth a look. Unfortunately, Madrigal didn't include any PCB artwork in their 'service manual' as most do. Option two is to only disconnect the VB during power off, which will certainly help. Option three is to leave it plugged into the VARIAC and turn down the AC voltage enough to keep the regs. regulating, but greatly lower the power dissipation. All 3T designs need a certain head-room from Vi to Vo in order to continue regulating. Option four is to use a small separate supply just to keep the memory powered 24/7, independent of the main. Like 'battery back-up'. Option Five is to reprogram my used input every time I use the main switch.

 

[sleddriver]

Update XII
After an extended period of Lifeus Interruptus, I thought I'd begin reconnecting the C7 into my main system. Utilizing the KISS principle, I started with only one source: DVD player. And to KISS it even smoother, I narrowed it down to video. In this case S-video.

I routed the S-video signal from the DVD player, through the C7, then back out via S-video into the V1 input of my aging 36" crt-tv. No joy. Furthermore, I wasn't able to display the C7's menu on-screen either via the same s-video connection. This is odd as I'd earlier tested this on another tv, but using composite instead. Unfortunately, the tv I tested the new VB on earlier is composite only. After searching for longer than I'd care to admit for the #$%! manual I'd "organized" somewhere, I was reminded that upon power-up, the factory default is composite video. Note the C7 only switches video of the same type. It's not capable of receiving a composite signal and outputing it as S-video, nor vice-versa.

So after setting the default to s-video, I tried again. No joy. Still wasn't able to see any of the C7's menu overlays either. I know they're being generated and outputed via the composite video buss. But they're not making it to or through the S-video buss.

So I removed the s-video cables and used composite video instead. Reset the C7 for all inputs to composite. No joy. This is odd. One complication is that I'm using a front-panel A/V input on the 36" tv because the 210# monster is 24" deep and sitting inside a custom entertainment center. Rear access takes quite a bit of effort to achieve. It might not accept another signal format as long as an S-video cable is plugged into the rear.

Nuts. Royal PITA....

Well one way is to start over using the C7 on the smaller tv, with no S-video, I originally used in a previous update. Enabling KISS, I used a single composite video cable between the MAIN composite video out on the C7 to a composite video in on the set. Power on...all looks fine...C7 main screen appears...default blue background, "No video signal" present appears (which is correct).

Alright...so as I go about connecting the composte OUT of the DVD player, the tv suddenly goes nuts...no more nice output. Note I hadn't yet actually connected the DVD player to any composite video input on the C7.

Everything was fine for a minute or so and then the wheels started coming off. Oscillation perhaps. So the NOS VB is now suspect, which is quite frustrating.

After turning the master power switch off for about 6 minutes, I turned it back on...Initial message came on...looks good....then within about 10s, the wheels start coming off.

Very odd...very odd indeed.

I repeat this three more times, with enough OFF time to allow all the EC's to drain. Same results. So something has changed between my earlier update and now and I'm not sure what it is. I'll have to check all the local video power rails for oscillations. Further, all of the EC's on this new board are at least 15yrs old now. Even if they haven't been in-use during that time, they're still suspect. After all, I spent a considerable amount of time testing and replacing all the EC's on the original board and verifed many were TU by testing them.

More later. . . .

 

[OVERKILL]

What an involved effort! Hats off to you sir for continuing on with this

 

[sleddriver]

Thank you for the kudos and for continuing to read my (now) very long thread! Murphy is alive and well in the soul of the Citation. Plenty of electronic landmines to navigate, defuse and remove. Some gremlin still persists however causing no joy. Sooner or later though he'll feel the heat of a 800°F soldering iron on his @$$ and I'll have the last laugh...and a cold one to toast my success.

Stay tuned!

 

[sleddriver]

Had some time late this afternoon and thought I'd take a look at the video outputs on the oscilloscope. They were all oscillating, both composite and s-video. So instead of taking a bunch of measurements, I decided to remove all three of the universal programmable array (UPA) IC's from the new board and install them in the old one I'd replaced all the EC's, local regulators and 100uH inductors.

This turned out to be the cat's meow! No more oscillations on the tv. C7 menu's were clean again. Next, I connected the composite video of a DVD player to In1 and it looked fine! Menu's looked fine. So I went through In1 - In8 and they all looked great.

Next I passed the audio through the C7, hooking the FRONT ouputs to the tv set and enabled STEREO ONLY. Audio was clear. Vol up/dwn/mute worked fine. No clicks or pops. I then checked all 8 audio input pairs and each was functioning. All good!

Next I took it into the other room, and used only s-video to the 36" set. On screen menu's looked fine, so I passed through some video, it looked fine as well. Checked all 4 s-video inputs and they are all OK. Also verified AUX & REC VIDEO OUT were functional.

It's been on now for a few hours and the video looks stable & clear. I guess the earlier testing of the new VB was too much for the old, dried out EC's. The power was on long enough to the board to heat up along with a few of the EC's and that was it. Their capacitance dropped enough to cause the wheels to come off somewhere and the 'new' board was now Tango Uniform.

So this is a relief....to say the least! It means I didn't make any mistakes replacing all the components on the old VB and that my trace repairs are all working fine. Further, it means the same for the much larger, more complex main board as even more EC's were replaced, along with the extensive trace repair around the PS.

Next, I'll hook up the center, sub & surround outputs, the amp trigger, recalibrate all the levels and program some settings and verify all is OK.

Given it's forecast to rain this weekend, I'm in for some (serious) surround movie time and seeing how I can tweak the C7 to add a bit of spaciousness to football.

time!