rca_14pd8053 Strange noise
 

I'm starting the process of restoring this TV. I have just finished a round of replacing several caps and have more to go. Testing my progress after some recapping, I decided to run the set on my isolation transformer/variac, and have noticed an unusual, loud and odd noise. Even with the sound all the way down, it doesn't affect the noise or its level.

I have a link to a clip I took. Sorry the crt is at an angle, as I had the other hand on the power supply and was reaching the other hand to take the video clip. The sound you hear at the end is due to the power supply being turned off.

https://www.youtube.com/watch?v=2flj...ature=youtu.be

I'm interested in what others think this is a telltale sign of or what it may indicate.

I am aware that the picture needs centering and the yoke needs adjusting.

The CRT tests and seems to appear strong.

Looking for any help, as I'm a noob.

Oh, here's the photofact as well, for reference.
https://drive.google.com/file/d/1QYI...ew?usp=sharing

Is the sound coming from the speaker? With that loud of sound you probably can feel the cone of the speaker vibrate. It could be that the vertical sweep signal is getting into the sound amplifier. Re-check your work. Some set use the audio amplifier to reduce the power supply voltage to other parts of the set. The electrolytic caps around the audio amplifier are very important. What you are hearing may be the result of changing a few caps at a time.
If you are unsure of where the sound is coming from, disconnect the speaker and substitute a 4 to 10 ohm resistor of a couple of watts size. It could be something else vibrating like the yoke or transformer.

That's one of those RCA's that use a dual purpose tube, 6CU8 for the IF and vert osc. Maybe there's something wrong with the tube.

Does the pitch of the sound change when you change the vertical hold knob?
I’m thinking that vertical pulses are getting into the audio output stage (since you indicate that the volume control has no effect)... likely bad filter caps in the 275 volt supply.

jr

Yes, in fact the sound changes as I adjust the vertical hold on on the back.
Good call. Most likely the remaining multi section filter cap and or the wax papers on the PCB, which still have to be done.

That sound is a bit intimidating. It sounds like the set is about to self destruct!

Will keep you guys posted and thanks for the help!

Bad filter caps in the power suppy are the most obvious problem. What caps have you replaced so far?

I've replaced filter caps and about to replace the multi section C3 (100, 20, 20) I've replaced all wax caps on the chassis facing the crt. Have not touched any caps on the PCB on the side yet.

C3 2 caps go to the audio section and one is in the power supply. So this is probably the main source of this noise

Update
 

Replaced C3 multisection. You can see its replacement at beginning of video, right behind CRT. Also made adjustments to yoke, vert lin and height. Previous strange loud noise has been eliminated. Vertical oscillation was definitely getting through the audio section.

Still need to go through the components on PCB board. Set is only passing audio and no video at this point.

https://youtu.be/bV2T3loJL_o

PS the plastic yoke cover has disintegrated and I had to remove it. The yoke is essentially floating on the neck of the crt. I'm thinking I'm just going to fasten a couple large plastic zip ties to hold it in place (?)

Here is a link to how someone rebuilt a flyback cover.
http://videokarma.org/showthread.php?t=268978&page=7

check your video detector diode and vid amp biases.

You mean the yoke cover? Very fancy. Does the cover serve any purpose other than a means of holding the yoke in place and acting as a sort of guide for the adjustment?

[QUOTE=Electronic M;3208469]check your video detector diode and vid amp biases.

Will check them out. In meantime I have to figure out how to work on the PCB now. Thing is impossible to remove without taking half the TV apart... No idea what the best way to work on this world be.

“ no video at this point.”

Does the brightness control have normal range?

jr

[QUOTE=Jon1967us;3208504] It's a hard one to work on, period!
I have two of those to work on, one 17" series string model and one 14" power transformer chassis. It's rather hard to believe that RCA would build a set that small with a transformer and a 5U4! I'd have to look at it to see if it's a two or three stage IF chassis, they built both. :scratch2:

Many PCB sets in the tube era were built with the intention that the PCB be serviced while still mounted to the chassis...Most offered some way to get at the foil side to solder, but some makes Like GE and Philco encouraged crushing the body of a failed original part, leaving the original leads as points to wrap/J-hook the replacement parts leads onto. I have service Lit for GE that literally recommends and instructs how to replace a part in that manner.

Funny you say that bc I considered doing the very same. I've been j hooking the ones where I can access the leads, but the radial ones with hidden leads proved daunting. I will give it a shot with some pliers.

In his case the rear cover also held the centering rings (these are magnetic).
The yoke has to be kept stationary with respect to the CRT because any movement between them will effect the image centering, size and rotation.

Seems to have normal range

Good.

Out of curiosity, what signal source are you feeding to the tv when you get sound but no video?

jr

Well, it seems that the detector diode is hidden in the metal can as indicated by the left most arrow. I have no way of getting to it other than spending 4 hours disconnecting and desoldering everything to remove the board. This board is totally unserviceable.

http://i67.tinypic.com/2qth3ja.jpg


Also, the Vid output tube has the test pins INSIDE the tube cover. No way to get to them with the tube in. The Mercury tube checker says the tube is fine by the way. Could I check the bias voltages with the tube out? Otherwise, there are barely visible traces by the socket. I may be able to check the voltages there.

Back to the detector diode, what are the chances it has failed? Do these old Ge Shottkeys have a reputation for failure?

Note, I am feeding a video signal in with a D>A converter box, connected to the antenna lugs.

I am feeding a video signal in with a set top D>A converter box, connected to the antenna lugs.

I will most likely secure these with large zip ties. Hopefully that will be sufficient.

They make a tool for getting those voltages with the tube installed... called a socket extender or socket saver....they go between the scoket and tube, lift the tube an inch or two and have test points on top.

I'd reckon video detectors prior to the late 60s had a 30% failure rate... the older the worse.
Most weren't shottkeys but Crystal diodes....they predated the transistor in consumer electronics.

Sample image. Many are available on ebay, get one that matches the number of tube pins you need.


http://www.videokarma.org/attachment...1&d=1550242038

I don't know exactly how your chassis is arranged, but sometimes getting to parts isn't easy, its just part of the task. When you have more restorations under your belt you get used to it and perhaps have a better eye for getting to things.
This article by Phil might give you some tips. Maybe in your case also it may make sense to remove the CRT to get to the detector diode.
https://www.antiqueradio.org/RCA_14-...Television.htm

Since the sound IF signal is taken from the plate of the video output, and the sound is ok, wouldn’t that imply that the video chain up to that point (diode included) is functioning ok? :scratch2:

jr

The diode rectifies the AM carrier to make video and passes the FM carrier for sound....if the diode shorts it won't make video for the video amp but the sound will still flow down the chain normally....heck if the diode opens and passes just enough signal through capacitive coupling to the video amp you could still get sound.... video detector fails are fun.

If you.can inject and watch video on the vid amp side of the diode then you know the diode failed.

Should be easy to test the diode under normal conditions, except that it's near impossible to get to. I can't really get behind the board to unsolder the can. Really thinking of a brute force way of getting to it, IE a Dremel or something :uzi: Would like to avoid spending upwards of 4+ hours unsoldering everything needed to free this board and then resoldering everything...but given the long weekend I just may have to bit the bullet.

I have plenty of germanium diodes on hand.

By injecting the video signal, could I plausibly take the yellow video out from the D to A converter output? would that be the right method? I've never done it before.

I'm aware of the Tube socket adapters. The problem here is that there is a metal cover, so one would need to pry that off to fit an adapter in.

OK, I didn't see the image when I first saw the post. I believe this configuration should work as it appears tall enough to clear the metal shield. I'll look out for one.

So there are no holes in the metal under the PCB to access the traces? I'm not familiar with that model, but many sets (even some members of the much maligned Philco Predicta family) have some holes under the board where with a really thin iron and the correct angle you can access most of the solder pads.

On sets where you can't get at the pads you need, if you are strategic on which wires you unsolder you can hinge a side of the board up something like 30 degrees to access all the pads by only disconnecting a small fraction of the wires.
Shielding is important in operation...If you damage the can you will come to regret it.

Test adapters are typically the same diameter as the thinnest tube that plugs into the socket so as long as it is taller than the shield there should be no problem using it...Techs needed their adapters to work ANYPLACE the tube would so if a company made one that didn't they'd get some harsh comeuppance from their customers. One thing you can do if you need an adapter fast and have a spare socket of the right type and a dud tube that fits it, is make your own adapter...Carefully bust all glass but the base, cut all leads inside the dead tube as far from the base as feasible and solder wires (3" is probably a good length) from them to the socket keeping the lead order correct.
A lazier way if you have several good spare tubes of the right type that you don't mind killing is to solder thin wires to the top of the pins for test points...If you heat too long the glass will crack and the tube will die of vacuum loss.

Video injection: With tube era test equipment like a B&K 1077 it is usually a matter of connect and done (the equipment has protection/isolation circuits in place to prevent damage).
With a modern signal source like a converter box plug and play can destroy your signal source unless you perform prior due diligence checks and use care.
First check: is your sets B+ and heater system powered through a transformer? If not you need an AC line isolation transformer to power the TV (and only the TV) thru before you can safely proceed.
Next (regardless of the results of the first check): is there a sizable AC or DC voltage between your sets signal ground and your converter boxes ground jack...If yes stop and get a TV service video generator or use a converter box that you don't mind possibly killing.
Now you can connect the outer shield of the converter box yellow video cable to the signal ground. Then use a .1uF 600V capacitor in series between the center hot lead of the video cable and your injection point....The capacitor couples the video signal to the injection point and blocks any harmful DC that could exist on your intended injection point (or from any adjacent point you might accidentally touch the lead to).

Also, bear in mind a converter box has standard video level and video polarity, and some TV sets do not use both at once at any valid injection point...That may result in an inverted monochrome picture, weak or overdriven video, and or no deflection sync (the synch separator relies on the right polarity and a decent approximation of the right level of video to function).
A proper TV test pattern generator for servicing has a video output that is adjustable for both polarities and amplitudes far greater than a converter can produce.
If you encounter a set with 2 video amp tubes the grid of the first may have negative video and the plate of the first (and grid of the second) may have positive video while the plate of the second has negative video again...Or every polarity is opposite. The waveform leaving the plate of a tube is always inverted polarity from that entering the grid. The polartiy that comes out the video detector diode is determined by the (even or odd)number of IF stages (as well as the correct polarity installation direction of the diode). So the number of IF stages dictates whether you get composite video standard polarity (but not necessarily amplitude) at the detector or not.

I have decided to tackle the board and try to disengage it from everything that's connected to it, so I can not only get at the diode, but take the opportunity to replace any remaining wonky components and check the traces, etc.

I don't have a video signal generator, so I'm going to have to rely on a conventional video source into the antenna input for now, but will try feeding into circuit in an alternate way, if I need to get to that point.

http://i65.tinypic.com/29q0gn6.jpg
http://i65.tinypic.com/j13wok.jpg

It looks like if you can get that bolt and soldered on under board shield off you can work on the board without removing the board.

Be aware if you plan to power it up before fully reassembling it that you will need to jumper any desoldered board ground to chassis. PCBs of this period usually did not have a contiguous ground plane, but instead had a few separate ground traces that each went to one of the board solder mounts...The board solder mount grounds effectively tied together thru the chassis. If any solder mount is disconnected or has a bad joint the portion of the board it grounds will float and the circuits that use those ground points won't work (until reconnected to chassis ground).

A few notes.
How do you service it ? Almost impossible ! My boss couldnt say
no to anything. There were still some floating around in the early 70's
and he refused to work on them. I dont know how RCA & others
built sets like this. Manuals said to cut out or crush then they had little
spring like things to make the connection.

Video detector diodes were troublesome up till the mid 60's. Often gave
odd symptoms so if any question just change the damn thing. When
built into a can often the can had a 2 piece shield & the top pulled off
so look for that.

BTW tube testers are almost a bigger liar than a politician. A known good tube
is the only way to be sure. We had a TOTL B&K & never used it. It was
for the customers & the biggest obscene profit maker in the shop.

73 Zeno:smoke:
LFOD !

I was finally able to work free the can hiding the detector diode. I desoldered one leg of the diode and tested it. It tested "normally" with a drop of .275 and no reverse current. I may replace it anyway, as there may be other problems with it undetectable by my multimeter and after 60 years of use.

http://i65.tinypic.com/wt81v9.jpg

http://i65.tinypic.com/n5ldoi.jpg

Another idea that ham friend, who is also very experienced in TV repair suggested, was to inject a 1Khz signal (using a .1 uF) along the signal path and see where there's a hard stop. The 1Khz signal should display black bars (according to him) on the screen. Injecting at a point where black bars don't display on the CRT is supposed to indicate the location of failed component.

I may see if I can get my hands on a tested/NOS 8AW8A to swap out.

Roger on the separated grounds on the board and making sure to clip those to chassis prior to testing. Great call.

The 1KHz signal may not work that well, unless you traced the signal with a scope. There would be no sync, so seeing bars on the screen would be very unlikely. You don't know how far off your sweep frequencies are free running.
That brings up a point, if you had no sync and the sweep frequencies were off, then a video signal getting to the CRT could have the image jumbled so much it would not be recognizable.

Have you had good luck doing this? I tried soldering to the leads inside a miniature tube and the solder didn't take. I believe the leads are nickle alloy or something like that which lead-tin solder doesn't work with. Maybe with a special solder, flux and a torch.

Works best with a square wave or pulses which are rich with harmonics... read here about the “mosquito” signal injector, a very useful device.

https://stevenjohnson.com/donbosco.htm

jr

I have had luck with this using normal 40/60 lead solder...Last year I pulled the 21AXP22 out of a CTC-5 (to stick in a rarer Moto that needed one) and replaced it with a glass tube...The change in CRT blocked the installation of the front/center tubes on the chassis. The easiest solution was to make an adapter with a 90-degree bend to lay the tube down on it's side on the chassis...I used the exact method quoted only bending the wires 90 degrees. It has worked fine for me so far.

Years ago (on my first resto) when I did not own a test pattern generator, but had an HP generator that covered 1Hz-10MHz I tried using that to inject into the video stages. I found that as I changed the frequency the number of bars changed from and they would vary between horizontal and vertical...If the amplitude is high enough and the video stages work you can get very contrasty patterns easily distinguishable from snow or blank raster...Also, sync is of little consequence unless your generator is fixed frequency. The sweep free runs fairly stable in most sets without sync so as long as you tune your gen to SOME multiple of one or both free run sweep frequencies you will get bars. If you don't tune it you get fingerprint-like patterns.
At some point in the ~10 years since I did that, I repeated that experiment on a different set (think I didn't want to dig the right generator out cause I'm lazy) and got the same results.