Questions about Bob A IF Alignment Series on YouTube

[Chris K]

I just watched the latest installment. Great stuff and very well instructed. I have a GE 805 "Locomotive" TV that I'm trying to align. Unfortunately, the instructions in the service data does not detail DC peaking with a VTVM and a signal generator. They show the sweep patterns to peak each slug to the correct shape of different traces they show. Can I use the method with the VTVM and just peak the coils or do I need to do it with the scope the way they instruct in the procedure and make little hills and valleys the right shape?

Thanks all.
Chris

[bandersen]

No, do it the way they describe with a sweep generator. Always follow the alignment instructions.

It is not a staggered tuned IF. Look at the schematic for the video IF. Notice there are no coupling transformers between stages?
It capacitively coupled with peaking coils. It's a different way of creating the desired IF response.

[DavGoodlin]

No surprise that GE used such a shortcut, even in the beginning, long before PC boards and compactron tubes.

[dtvmcdonald]

Originally Posted by Chris K

I just watched the latest installment. Great stuff and very well instructed. I have a GE 805 "Locomotive" TV that I'm trying to align. Unfortunately, the instructions in the service data does not detail DC peaking with a VTVM and a signal generator. They show the sweep patterns to peak each slug to the correct shape of different traces they show. Can I use the method with the VTVM and just peak the coils or do I need to do it with the scope the way they instruct in the procedure and make little hills and valleys the right shape?

Thanks all.
Chris

Sure you can just use a voltmeter and a calibrated signal generator.
But no matter which version of the set you have, you need a bias
box (a 9v battery with a 5 or 10K pot across it) to set the bias voltage
at the junction of R241 and R243 to -4 volts. This is because the DC from the output otherwise backs up to there ... this does not matter at 60 Hz.

Use an analog meter between pins 2 and 3 of V8A.

Then you just sweep the generator slowly by hand to get the patterns shown. Tedious but it works.

[Chris K]

Originally Posted by bandersen

No, do it the way they describe with a sweep generator. Always follow the alignment instructions.

It is not a staggered tuned IF. Look at the schematic for the video IF. Notice there are no coupling transformers between stages?
It capacitively coupled with peaking coils. It's a different way of creating the desired IF response.

Yes I see that now. Thank you! Unfortunately my B&K 415 only puts out about 30 MHz at the lowest IF output setting and this requires 25.0 as the center point. I wouldn't be able to use the markers either...I'd need an external marker unit. I do have a decent Knight frequency generator that I have dialed in at 25 MHz using my scope but I can't sweep. I do have a Knight sweep generator I restored but it's electromechanical and I would need to calibrate it first.

[Chris K]

Originally Posted by dtvmcdonald

Sure you can just use a voltmeter and a calibrated signal generator.
But no matter which version of the set you have, you need a bias
box (a 9v battery with a 5 or 10K pot across it) to set the bias voltage
at the junction of R241 and R243 to -4 volts. This is because the DC from the output otherwise backs up to there ... this does not matter at 60 Hz.

Use an analog meter between pins 2 and 3 of V8A.

Then you just sweep the generator slowly by hand to get the patterns shown. Tedious but it works.

I might try this if only for the exercise. My B&K 415 has a DC output

[Electronic M]

Originally Posted by Chris K

Yes I see that now. Thank you! Unfortunately my B&K 415 only puts out about 30 MHz at the lowest IF output setting and this requires 25.0 as the center point. I wouldn't be able to use the markers either...I'd need an external marker unit. I do have a decent Knight frequency generator that I have dialed in at 25 MHz using my scope but I can't sweep. I do have a Knight sweep generator I restored but it's electromechanical and I would need to calibrate it first.

There's an old thread here on Videokarma where someone with a B&K 415 found cheap modern RF hobby oscillator boards and came up with a way to externally heterodyne the 415s output down to the old 21MHz TV if standard and use it to align early sets. It might be worth digging up.

[Chris K]

Originally Posted by Electronic M

There's an old thread here on Videokarma where someone with a B&K 415 found cheap modern RF hobby oscillator boards and came up with a way to externally heterodyne the 415s output down to the old 21MHz TV if standard and use it to align early sets. It might be worth digging up.

I'll look Tom. Thank you

[bandersen]

Originally Posted by dtvmcdonald

Sure you can just use a voltmeter and a calibrated signal generator.
But no matter which version of the set you have, you need a bias
box (a 9v battery with a 5 or 10K pot across it) to set the bias voltage
at the junction of R241 and R243 to -4 volts. This is because the DC from the output otherwise backs up to there ... this does not matter at 60 Hz.

Use an analog meter between pins 2 and 3 of V8A.

Then you just sweep the generator slowly by hand to get the patterns shown. Tedious but it works.

You set the contrast control to fix the AGC bias at -4 volts. No need for a battery or external supply.

Tedious? That would be insanely difficult. Rotating the RF gen over about 5-10 MHz, while simultaneously looking at a meter needle swing and translating that into a graph in your mind? While keeping careful note of the response at key frequencies and tweaking coils while doing that?

Get a cheap sweep gen. You can even use the TinySA in sweep mode.

[bandersen]

Here's a demo using the TinySA as a sweep generator. They don't bring the x-axis sweep voltage out of the device, but it is possible to pop it open and add a wire.
Otherwise, set your scope to a slow time base and play around with the triggering to get a stable display.

https://youtu.be/BQJjgpHyWRs

[dtvmcdonald]

Originally Posted by bandersen

You set the contrast control to fix the AGC bias at -4 volts. No need for a battery or external supply.

Tedious? That would be insanely difficult. Rotating the RF gen over about 5-10 MHz, while simultaneously looking at a meter needle swing and translating that into a graph in your mind? While keeping careful note of the response at key frequencies and tweaking coils while doing that?

Get a cheap sweep gen. You can even use the TinySA in sweep mode.

Its actually not difficult. I've done it many times, though using a DC oscilloscope makes it easier to see a curve. The problem with the scope (at DC) is that often you need two probes and two subtracted inputs. I have a pile of Tektronix scopes.

And yes, you need a bias box. You don't for a sweep generator because there are capacitors that filter out the AC so you dont get that acting as AGC. But at DC more or less, those caps are too small.

I used this method for TVs with IFs at about 10 MHz before I got an SDR sweep generator working correctly. Yes, back at the dawn of time there were such sets, and they still work.

[Electronic M]

I've heard of what's being described by Bob and Doug sweeping an input, logging and manually graphing the meter readings and response curve. I've heard of it being used by engineers in the tube era.
For either a very low resolution graph skipping large frequency swaths, a more simple system than a TV IF, or if the world comes to an end and all the few survivors have to work with for a reading is a simple meter I could see this being viable. But for a TV alignment it seems like as fun a task as trying to build a decent sized modern stone building using only the tools and methods of the ancient Egyptian pyramid builders from quarrying through construction...Sure you can do it, but the time a modern tool saves should more than pay for said tool.

[bandersen]

No, you don't need a bias box for a GE model 800. The contrast control sets the IF gain. There is no AGC. It doesn't work like a conventional design and takes some getting used to.

And, yes, you do need a fixed bias box when sweeping for other designs.