Welcome to Australia's only Vintage Radio and Television discussion forums. You are not logged in. Please log in below, apply for an account or retrieve your password.
Australian Vintage Radio Forums
  Home  ·  About Us  ·  Discussion Forums  ·  Glossary  ·  Outside Links  ·  Policies  ·  Services Directory  ·  Safety Warnings  ·  Tutorials

Tech Talk

Forum home - Go back to Tech talk

 Oscillation in STC 830 radio
« Back · 1 · 2 · 3 · 4 · 5 · Next »
 Return to top of page · Post #: 31 · Written at 1:30:19 AM on 16 February 2023.
Marcc's avatar
 Location: Wangaratta, VIC
 Member since 21 February 2009
 Member #: 438
 Postcount: 5461

The chopstick method tends to indicate that the problem is isolated to the front end. One of the reasons I do an IF calibration first is that the measuring instrument is normally an Oscilloscope. Commercial fixing for decades has taught me that a Signal generator & Oscilloscope are a formidable combination for analysing signal path faults.

Even with no Oscilloscope if you try calibration and it fails you go back to the last IF stage tube inject & calibrate and attempt to progress toward the Pentagrid. If it works all the way but not when you get to the Pentagrid /Mixer. Then its that section where the fault is.

The injection of too much signal, will cause the AGC to kick in and you will never calibrate it. That is why they suggest high volume & low signal like Microvolts and I originally used pF series caps in series with the Signal generator lead.

Calibration of the Generator is also important, mine is a entry level job and bit rough & tumble with the dial calibration. That's why I use a Frequency Counter (Fluke) to calibrate the generator. Traps: calibrating with the modulation / tone "on" and unhooking it.

As I have repaired a lot of radios, the audio of the generator is used to sync an oscilloscope via an earthed metal box, with semi-permanent connections & a two stage fixed attenuator in it.

As Fred has intimated: With fixing, it demands a logical and methodical approach starting at the power point and working toward the antenna Many Australian circuits & American ones are scant on detail with element voltages, but they are critically important.

Marc


 
 Return to top of page · Post #: 32 · Written at 3:57:25 PM on 16 February 2023.
STC830's Gravatar
 Location: NSW
 Member since 10 June 2010
 Member #: 681
 Postcount: 1306

Fred: Have ordered Silicon Chips micromite sweep generator. Until that turns up I am looking at items that need attention like shielding and the padder trimmer. So you are right, I am diving about a bit, but hope to get down to a more sytematic investigation before too much longer.

As far as calibration goes can I assume that the micromite gear is accurate?

Marcc: Thanks for the detail, I think that you have covered this before eg your Coupling box & attenuator thread in Workshop Tools and Test Equipment forum.


 
 Return to top of page · Post #: 33 · Written at 5:22:10 PM on 16 February 2023.
Marcc's avatar
 Location: Wangaratta, VIC
 Member since 21 February 2009
 Member #: 438
 Postcount: 5461

True: One would like to think others have looked at that forum, seen it, evaluated it and perhaps even built one? The objective is to demonstrate ways to expediate fault finding. Everyone has there own ideas on that & I have seen results from some who have no idea.

Its to a point of "What works for you", While providing ways to proceed with the highest probability of success.


 
 Return to top of page · Post #: 34 · Written at 7:47:25 PM on 16 February 2023.
Fred Lever's Gravatar
 Location: Toongabbie, NSW
 Member since 19 November 2015
 Member #: 1828
 Postcount: 1331

Hi STC, I simply use the unit as a sweeper source of signal.
You can set the sweep range by the touch screen to 6 places.
The accuracy of the unit is tied to the internal frequency reference and gives comparable figures to my frequency counters.
Check out my post on it in "special builds".

The designer uses it as a combined sweeper and readout on the screen.
That just confuses the crap out of me.
Someone else who has used the device may like to comment.
I wanted to see the actual sweep shape or profile so use my CRO hooked to the detector output.
Of course, you can also sweep audio circuits or anything within the frequency range setting.
A very handy unit.

At least at the end of all this your knowledge of the superhet radio will be given a good boost!

Fred.


 
 Return to top of page · Post #: 35 · Written at 11:45:20 PM on 16 February 2023.
Marcc's avatar
 Location: Wangaratta, VIC
 Member since 21 February 2009
 Member #: 438
 Postcount: 5461

It is always the set that does not play ball, that is the educational one. It forces you learn. If you learn how the thing works: You have a better grounding for working out why it doesn't.


 
 Return to top of page · Post #: 36 · Written at 2:34:16 PM on 17 February 2023.
STC830's Gravatar
 Location: NSW
 Member since 10 June 2010
 Member #: 681
 Postcount: 1306

Fred & Marcc - I agree, hands on best for learning.

I wonder why they used a log vertical axis - maybe the limitations of the small screen or else the marks indicating measurements wouldn't show clearly enough. Will find out soon enough.

Is the plastic jiffy box adequate given the need to avoid signal leaking to where you don't want it?


 
 Return to top of page · Post #: 37 · Written at 8:04:47 PM on 31 March 2023.
STC830's Gravatar
 Location: NSW
 Member since 10 June 2010
 Member #: 681
 Postcount: 1306

Finally got back to sorting out this problem. Have built the SC micromite signal generator with alignment facility. The first thing I did with this was to feed 1000Hz modulated medium wave signals into the radio starting at 1500Kz, and discovered that MW does accept the full MW frequency range with decreasing sensitivity down to 550kHz, but they are displaced on the dial with 1500kHz where it should be, and 550kHz appearing at about 700kHz on the dial, just before where the oscillation starts. So there is a serious problem with alignment with regard to tracking. First thing I did was check the padder capacitor, but the variable section works OK and the parallel mica is very close to spec.

Next applied signal to SW2 which is 2.8 to 7.9 MHz, and got a signal through the full span of the dial, in close to the right locations. So since SW2 is working it means that signal is getting through and the RF and IF sections must be working over the full range of tuning gang movement; so the MW problem can't be related to a faulty tuning gang.

A complication is that I cannot receive a signal up to 10mHz (the upper limit of the signal generator) on SW1 (7.6 - 24MHz). However it picks up noise and some stations so is working. So this issue can wait.

So it seems to me that oscillation of MW is related to other than IF transformers and is centred around 6A8 mixer. So will look closely at circuit around 6A8, being MW RF and oscillator sections. Also will look at response of individual IF transformers, particularly IF3 which doesn't align properly as discussed in post 14.


 
 Return to top of page · Post #: 38 · Written at 1:00:22 AM on 1 April 2023.
Marcc's avatar
 Location: Wangaratta, VIC
 Member since 21 February 2009
 Member #: 438
 Postcount: 5461

The IF and oscillator are not related for the purpose of alignment of the IF. I don't know if the Micromite has DC blocking but it is important to not have a resistive coupling the the signal grid (Top cap of the 6A8) and you only need around 50uV into the signal grid as probably noted. If the AGC cuts in, it will never set.

AGC looks like its applied to several tubes so switching to SW may not save you. Just keep the input down & the volume up. 6U7 is a horror tube for destabilising if not shielded properly. One gets the IF right before messing with the front end. One hopes that no Mica caps were changed on coils.

The first step for the front end is setting the mechanical tracking of the pointer and gang. As that has an RF (closer to TRF) stage ahead of the 6A8 in order to function applicable to all bands the aerial coils on the input of the TRF stage, must be on the same frequency as the ones feeding the 6A8 signal grid. Oscillator should lead the radio station (normally) by the IF frequency plus the station. e.g. 1000kHz station plus 450 = 1450kHz That can be picked up by a transistor radio in close proximity.

Do make sure that the tuning gang is not shorting, that being from debris corrosion & bent plates. If you cleaned the Mica Trimmer and used a circuit board cleaner; make sure that there is no moisture.

Where there is a padder C16? that is normally set at the low end around 600kHz and trimmed at 1400 kHz with the trimmer.

One must be very careful changing Mica Caps in the front end. The coils & gang are a factory matched set, interference with them can have serious ramifications with band spread etc.

With fraying cloth covered wire, trim it and, paint it with glue, or circuit board lacquer, or use shrink tube. lacquer should really be use on joints of dissimilar metals.


 
 Return to top of page · Post #: 39 · Written at 2:18:28 PM on 1 April 2023.
STC830's Gravatar
 Location: NSW
 Member since 10 June 2010
 Member #: 681
 Postcount: 1306

Thanks for comments Marcc

The Micromite has a 600V poly capacitor in its RF output for DC blocking, and RF output is adjustable on a 1 to 100 scale.

So far I have just fed modulated MW into the front end because I previously observed that SW1 & 2 were free of oscillation and picking up stations and I wanted to see what their tracking was like. Good tracking on SW2 means to me that the tuning gang is innocent of causing MW oscillation or poor tracking at the low frequency end. The fact that I can't get anything below 10MHz (the upper limit of the generator) means tracking of SW1 may be out, but as I said is working over its full frequency range, so again the tuning gang innocent. So it should be as when I reinstalled on the chassis after cleaning and painting before this trouble started, the MW tracking was perfect. So I think that the mechanical tracking of the pointer and gang is good.

AGC is applied to RF and IF valves, not mixer. Only mica caps messed with are C5 in the filter network on the output of IF3 transformer going to the detector 6H6, and C14 across the padder C16. But again before present troubles, and C14 very close to spec. After cleaning C16 with isopropyl alcohol heated in the oven at 100C for an hour to drive off any moisture. C16 0-0.12nF, C15 C16 combination 0.29-0.42nF. I suspect C13 the SW1 padder because tracking appears to be out.

Padder C16 is screwed right out in an attempt to get alignment at the low end but what should be 600kHz appears at ~700kHz. Something is throwing that out before the oscillation appears slightly lower than 700. 1400 comes in nicely with the oscillator trimmer.

Treating frayed wires has happened if they needed to be lifted; the rest of these ends are the myriad little jobs needing doing. Just finished resheathing the speaker cable with braided sheathing, heatshrinking frayed areas along the length of these wires, and discovered that one of the leads to the choke ie HT had only two strands left intact where clamped to the speaker. New piece soldered in and heatshrinked twice.


 
 Return to top of page · Post #: 40 · Written at 12:44:01 AM on 2 April 2023.
Marcc's avatar
 Location: Wangaratta, VIC
 Member since 21 February 2009
 Member #: 438
 Postcount: 5461

It is more common to get parasitic oscillation with modern caps mainly in the area, as noted, around the 2nd detector 1st Audio and that when lead dress, & shielding gets critical. I have often had to add shielded wire into that section.

You could have a parasitic oscillation from an unstable stage and that can drive the AGC high. Worst offender 6U7 aka 6D6 (different base only).

The other thing is that internal radiation in the set will be strongest at the lower frequencies as the pentagrids tend to lose modulation at the frequency increases. Indicating that a problem like yours is likely shielding & lead dress. Things like wires need to be kept close to the chassis.

If you move things around in the IF can the wires, close to walls, are a capacitor.

Marc


 
 Return to top of page · Post #: 41 · Written at 8:34:36 PM on 5 April 2023.
STC830's Gravatar
 Location: NSW
 Member since 10 June 2010
 Member #: 681
 Postcount: 1306

Going by your Post #9 I have attempted to check IF stage resonance.

The oscillator was shut down by shorting out of the the primary of the MW oscillator transformer. Shorting the secondary wasn't totally effective because some broadcast signal was still getting through.

The signal generator set at 450kHz was connected to the grid of the 6A8. No amount of signal delivered by the generator would shift the AGC voltage above about 0.6volt. AGC connection point for the meter was for the two IF valves (the other part of AGC is delivered to the RF valve via a radio frequency choke). Despite the low 0.6volt AGC still managed to peak all of the IF setings except the secondary of IF3 connected to the detector valve, the adjustment screw of which is wound right out to get maximum reading. So will see if there is something causing an increase in capacitance across this winding such as contamination of the trimmer micas.

This alignment decreased the sensitivity of the radio as far as normal MW operation is concerned with generator disconnected and aerial connected and oscillator unshorted, but did stop the oscillation at around 650kHz. It seems to me that alignment as described here produces a different result as far as sensitivity is concerned, from alignment by ear of a weak signal fed in through the aerial.

This difference may be due to the low AGC voltage so will have a look at applying an external voltage to the AGC. Or else I have done something wrong.

Any other ideas gratefully received.


 
 Return to top of page · Post #: 42 · Written at 9:11:06 PM on 5 April 2023.
Fred Lever's Gravatar
 Location: Toongabbie, NSW
 Member since 19 November 2015
 Member #: 1828
 Postcount: 1331

OK.

1/ the IF3 tranny:
It is screaming at you it has a problem!
Find out what it is!

2/ any decent set will generate 30 to 40 volt AGC OPEN CIRCUIT which when looped backto the grid circuits will tend to make the front end pivot around a value perhaps between say 5 to 15 volt..
That what AGC does!

3/ You say AGC is 0.6 volt.
That is just so wrong.
There is something very amiss here.
If 0.6 volt is the only AGC voltage , all the front end tubes will be running "full throttle" and you should have 40 plus AGC volts!!!
EXCEPT, full bore tubes may be running at saturation with bugger all amplification, so the thing just chokes down!
That all points to something very wrong around the detector and AGC diodes and the associated IFT.
Low AGC means a heavy load or an open circuit on or to the AGC diode.
The AGC system is not there for fun, on a potent set with 3 or 4 amplification stages it is VITAL the tubes are working on a straight part of the mu curve, not saturated and not cut off.
You must have the AGC system working properly so the front end rides up and down the amplification slope to give AUTO MATIC VOLUME CONTROL on different signal levels. Sorry about the shouting!

Forget trying to "align the set"
Make the front end work properly by fixing the AGC problem!

Fred.


 
 Return to top of page · Post #: 43 · Written at 12:39:43 AM on 6 April 2023.
Robbbert's avatar
 Location: Hill Top, NSW
 Member since 18 September 2015
 Member #: 1801
 Postcount: 2109

Make sure the secondary of IF3 is conducting. I've had them go open-circuit and the radio still almost works, even with the diode effectively disconnected. Or perhaps it has shorted inside the can. But the fact that the slug is in the wrong position tells you there's a large problem there.


 
 Return to top of page · Post #: 44 · Written at 8:36:34 AM on 6 April 2023.
STC830's Gravatar
 Location: NSW
 Member since 10 June 2010
 Member #: 681
 Postcount: 1306

I agree 0.6V AGC is so wrong; can get more normal AGC voltage feeding MW signals into the front end when tuning by ear. So I think I haven't set up the 450kHz to 6A8 grid/ short the oscillator method properly somehow. I pulled this set-up out and came back two days later and set it up from scratch again with the same result.

Secondary IF is 10.2ohm both primary and secondary so it is definitely conducting. Will investigate (again) all the connections and components between IF3 and 6H6 detector.


 
 Return to top of page · Post #: 45 · Written at 7:48:42 PM on 6 April 2023.
Fred Lever's Gravatar
 Location: Toongabbie, NSW
 Member since 19 November 2015
 Member #: 1828
 Postcount: 1331

STC forget the 6A8 set up thing and the response shape.

Just run it like a normal radio and inject a MW AM modulated signal into the aerial terminal( say 1000Kc) and tune it in.
The mixer will produce a IF signal for you.
Peak the IF's to this IF signal.
If normal you should have a AGC that changes as you change the input signal level.

If the IFT3 still does not peak, find out why.
Delete that IFT if necessary and run with 2 IFT's with IFT2 driving the diodes.
The set will still work happily that way with just a lowered sensitivity.
If now the IFT2 wont peak, something in the diode circuit is loading it.

You must get logical operation of the front end before you precede.

Fred.


 
« Back · 1 · 2 · 3 · 4 · 5 · Next »
 You need to be a member to post comments on this forum.

Sign In

Username:
Password:
 Keep me logged in.
Do not tick box on a computer with public access.