Six Weeks of Troubleshooting...I think I found it!
 

Before Thanksgiving, I got stalled trying to align the sound and video IF traps on a 1949 RCA KCS 28B chassis in an RCA 9-TC-240 cabinet TV. After making some progress (With a tremendous amount of help from Penthode, Tom C and Bob A) I began to have issues with the HP 8600 Sweep Generator and the HP 8601 Digital Marker combination. Without describing the long ride to standstill, and after several "Ah Ha!" moments that I thought solved the issue, I all but gave up after the last failure. The furthest I got was narrowing the issue down to the 8600 marker. Sometimes the issue would disappear with banging or circuit board flexing only to show up an hour later or a couple of days later. I figured it was a ghost in the machine I'd never find.

Six weeks of actual troubleshooting using a multimeter and a schematic instead of a club led me to the 5V circuit on the regulator board. It was low at 4.5V even with the adjustment pot on the board set for maximum voltage. The voltage is regulated by a 10 gold pin HP can IC, part #1820-0196.
It's proprietary...nobody but HP made it. When I froze it, voltage went to +5V DC. The machine worked perfectly until the voltage returned to 4.5V and the problems reappeared. I then froze it to restore function and instead of waiting for it to warm up, I immediately adjusted the voltage down using the regulating pot and as soon as I got to 4.5V, the marker wouldn't function correctly. I repeated this several times and was able to reproduce the phenomena every time. The regulator is a 500 ohm pot and it measures perfect. Below is the offending board and the SOB of a "chip in a can" part.

https://imagizer.imageshack.com/v2/1...922/GcQU2p.jpg

https://imagizer.imageshack.com/v2/1...924/Wm5E76.jpg

https://imagizer.imageshack.com/v2/1...923/P84ZUo.jpg

After fruitless searching for weeks, I finally came across the chip NOS on eBay for $10. So it should be here in a few days , I'll instal and test the unit. If it's now reliable, it's back to alignment on the RCA.

While thankful for all of the contributors in the other thread (and there were mare than the 3 I mentioned), I want to give a huge thanks to Penthode who rode with me through this half year process giving advise and imparting his experience and wisdom while taking a heartfelt interest in seeing me succeed. Besides the thread we exchanged emails as well. I hope he'll be pleased I have not given up!

Additionally, I bought an HP 8643A Signal Generator. It's a beast. Other than a dead memory battery, it was tested and certified and appears to work well. I can sweep with it but if I'm going to look at a scope XY trace, I need to figure out where to tap into the generator to access the horizontal sweep output. The scope gets the Y input from the test point in the TV, the sweep signal goes into the input lug of the particular tube but the scope needs an X input. It's probably on the inaccessible back of this station wagon on my bench!

Thanks everyone!

Regarding HP8643 - beast is an understatement! You'll find the x-axis output on the back of the device

Of course it is! Thanks Bob.

Nice work Chris getting that debugged. You sure don't run into those early IC's to often these days. When you get it running it will be a hell of a nice sweep generator.

I have a sync board in my GBC small studio video cameras that has Faichild epoxy dome IC's that are sort of like these. There's about 12 of them and thankfully they're all good.

The HP8643A is awesome too with 3 programmable markers.
https://www.machine--tools.com/By-Lo...ator-img-2.jpg

I got the X axis hooked up and its works great but on Auto Sweep I can't get it to do more than one sweep per second. I can see the cursor trace and it looks like it's supposed to but... Can it sweep faster than that? I don't think it's useable like this.

Thanks. I'm a very inexperienced tube radio and TV guy. Solid state isn't something I'm comfortable working on. Upside is there's no 350V B+ to getcha!!!

The HP8643A is a terrific piece of equipment. But it is a bit of an overkill for this application. I hope it works for you.

Hey there you are!!! Welcome back to my adventures! Yes, overkill for sure and once the 8600 is up and running, I'll be using it and the 8601 but I wanted to see what this could do sweeping and I'd like to try a bunch of other stuff with it too. I have vintage signal generators but accuracy and reliability of the analog stuff has led me down a lot of unnecessary rabbit holes. This piece of equipment should be all I'll ever need.

HP 8643A supports 20 ms to 10 sec sweep times. Press sweep time to set it

Yes I see that in the sweep range limits but when I try to enter it I get an error message "Sweep time too small". I have a 1MHz sweep with 21.25MHz as the center frequency. Something else is wrong. It's set to the default phase continuous sweep.

What value are you entering? I think it's in seconds so 0.1 should give you 100mS sweep.

I’m concerned it can’t do it. If you look in the table it says if your a range where 21.25MHz sits, minimum is 0.5 seconds. I tried to enter 0.1 seconds and got the same too small message. When you see a MHz frequency displayed on your HP of this type, how many digits to the right of the decimal does it display? My display has about 8 zeros.

Use phase continuous mode. It is enabled by activating special function 112

Yes I've been going to that. It tells me my frequency span is too large for phase continuous. I'm scanning 500KHz to either side of the center frequency. What the hell? Something basic in my setup can't be right but the trace on the scope looked fine. The sweep just wasn't fast enough for me to see the scan as a line.

https://live.staticflickr.com/65535/...b50bde8d_c.jpg

https://live.staticflickr.com/65535/...c3aa198c_c.jpg

https://live.staticflickr.com/65535/...65963a08_c.jpg[

https://live.staticflickr.com/65535/...4682d0df_c.jpg

Does your unit display MHz values like this? What's the purpose of showing out to more than 3 decimal places?

https://imagizer.imageshack.com/v2/1...924/Zq2bQj.jpg

Did the steps exactly as you did. When I get through the sweep time and push auto sweep I get the too large span message. Same too large for single sweep and manual says it can't do it in continuous mode...obviously

There's a special function I saw for wiping all saved setups clean. Maybe I'm in some saved mode that's stopping me from doing this...IDK

So you press special then 112? I don't know what to tell you

Yeah...I know. Here we go again with this crap.

I started with auto, manual and single all off. Entered special 112, then center frequency, then span, then rate then turned auto on.

Can you post a photo showing your entire control panel?

The widest sweep span it’ll do if I ask it to do 100ms sweeps is 50KHz. I pretty sure somewhere in the documentation for this it gave center frequency ranges in increments of 20MHz. I think it said from 16 to 30 or so MHz the fastest it would sweep is half a second. I have no idea why that would be the case. The fastest it would do a 1MHz sweep around a 21.25 CF was 1 per second. That’s as fast as it could go. If I lowered it below a second I got an error saying the sweep interval was too small for the frequency. Looks like my future lies in the older HP generator/marker and the effectiveness of my replacing the chip. If the beast can’t sweep it’s gone. My equipment winning streak continues!

Sure thing but I’ll need to do it tomorrow evening. The work week becons. Thanks for the help tonight!

You are describing the behavior in digital sweep mode. Phase continuous is a smooth linear sweep and recommended for IF work.
Also, the slower you sweep, the more accurate the results are. Is 0.5s really so bad?

https://live.staticflickr.com/65535/...527a8556_o.png

Oh 1MHz sweep is too wide. Lower it to 0.625 MHz or less
https://live.staticflickr.com/65535/...3df3df7f_o.png

You are describing the behavior in digital sweep mode. Phase continuous is a smooth linear sweep and recommended for IF work.
Also, the slower you sweep, the more accurate the results are. Is 0.5s really so bad?

I guarantee you I am in phase cont. mode. Maybe I'm missing a solution with the wrong machine. My issue is not with the quality of the sweep but with what is visible, or should I say not visible, on the oscilloscope. I don't have a curve shape on the scope, just a cursor that traces the curve shape once or twice a second as per how fast the HP is sweeping. I'm trying to make adjustments in the TV IF transformers that alter and tweak the symmetry of the curve around 21.25MHz on the scope. With the previous 8600/8601 combo, I had a sweep rate that was so fast, I had a solid line on the scope out of the TV so I could make fine adjustments to the TV IF transformers that got the shape exactly right. Is there something I should be setting the scope for...some mode that will display a line instead of a slow moving cursor?

You need to decrease your sweep width. You can't do 1MHz - 0.625MHz or less.
A slow moving cursor is all that you will get at 0.5s sweep rate.

No, this isn't the best generator to use for TV alignment. It's one of the finest RF generators ever made, but it was designed for communications work and cost a small fortune.
8662A would have been a better choice. Older, but has 5 markers and less restrictions on sweeping. I don't think it has a continuous mode though. Only discrete steps so not as nice in that regard.

OK Bob. I’ll keep it for now for other uses but it seems Penthode is right. The 8600/8601 pair is the sweet spot for this era of TVs. Thank you for all the help and your time!

Well at least try reducing the sweep width to 0.5Mhz to verify your generator is working properly.

Okay will do.

It worked at 0.5MHz.

Better news is the voltage regulator chip arrived and is installed. The 8601 is working. Five volts DC on this DC circuit now. Fingers crossed it stays stable.

https://imagizer.imageshack.com/v2/1...922/trqxir.jpg

Congratulations on finding the bad regulator and hopefully it solves your issue. The HP 8601 along with the HP 8600 Digital Marker combination are most likely the best relatively modern thing going for TV alignment. If you feel challenged while troubleshooting those two, you'll be horrified of the 8664's internal complexity, as well as the 8662. They are not at all user friendly service wise.

It would appear the HP 1820-0196 is a Ua723 regulator in a TO99 package. A fairly common part from that time period.

Yes the overall challenge is what I am afraid of. There is also a factor of trust with the equipment as the complexity removes you from the actual task.

The main pros of the 8600a/8601a are multiple accurate markers, marker adding does not interfere with the displayed frequency response and simple sweep and frequency and amplitude controls.

Yes further research showed that. You can still get them with the right form factor for a bit more than I paid..DigiKey and Mouser among others.

Lesson learned! It's been working great so far and I was able to produce nice traces on the scope for the discriminator and sound response traces for the RCA.
Next is getting to these traps...again. I found and serviced the exact RCA VoltOhmyst VTVM from this era recommended for the alignment. I had some custom probes made up by an eBayer who is well known for quality reproductions of old test equipment probes and leads. What could go wrong...right???!!!

You need first to make sure before adjusting the traps is ensure the IF adjustments are roughly correct. I would suggest first assessing the state of the VIF stages as I believe you may have inadvertently adjusted two stages to the same frequency which may indyuce oscillation.

When adjusting the traps, you may have to increase the input signal and/or reduce the bias to get a clear indication on the VTVM. With this alignment you have to check, recheck and check again that you have not adjusted something to an odd spurious peak or null. You have to moderate the IF amplifier gain and input signal to it to ensure the amplifier is linearly amplifying. That is not too high and not too low. The signal output level of the amplifier at the 2nd detector is important to constantly monitor and reference.

I have continued on about this because earlier (aside from the equipment deficiencies) this issue I believe caused you no end of grief earlier. The RCA Alignment instructions are only a guide. You must understand what you are doing and stop if you have any question.

Again, before adjusting the traps, you must ensure the IF amplifier is roughly correct and that no oscillation ensues when you reduce the control bias.