I recently bought my first boatanchor and have recapped it and tested and replaced several tubes. The AVC function does not work and after checking voltages on each tube socket terminal I find that the grid voltages on the 3 tubes associated with the AVC are low( the tubes are RF AMP, 1st@2nd IF AMP). Example RF AMP factory grid voltage, -1.8VDC; actual -.25VDC. The AVC switch works.
Also the 6H6 DET,ANL,AVC has low voltages while other tubes have close to factory specs. The set receives stations on all bands and audio output is good, while the S-meter only works on very strong local AM stations. Does anyone have any thoughts on what could be wrong with the AVC?

Chuck

Trace the circuit carefully, chances are a miss-connection was made with the caps or resistors...

Your meter may be loading down the grid voltage readings.

Re-check the alignment, be sure it agrees with any crystal filter it has or has not...

Try another 6H6. AGC current is almost infinitesimal, any leakage from rotted rubber wires or some sort of tuner cleaner residue on switches will kill it as well as a tube with a poisoned grid.

Chas
WA1JFD

Check resistors. Many resistors in older Halli sets have drifted 1.5 to 2x marked value.

--Chuck

The fact that the "S" meter only works on strong stations may also be a good clue, since those would also give the best AVC voltage.

I agree with the earlier comment about checking the alignment. Even though you may think audio output is adequate, it might be out of alignment.

Check the screen voltage on all those tubes, a changed dropping resistor can kill the gain.

Carl

Thanks to those who have responded to my AVC problem. Are there any current or former Hallicrafters SX-99 owners who may have additional suggestions on this issue?

Suggest you replace C22, C33, C41 and C42 if you have not done so already. These are mica caps, which are usually more reliable than paper caps - but there have been reports of the mica caps in Hallicrafters IF transformers from earlier radios failing.

Also, suggest you measure the resistance from the AVC line to ground. Across C58 would be an easy place to access it.

--Chuck

Check the 6.8 ohm resistor on the filament of the 6H6 (note: the Sams is drawn incorrectly).

N7RHU--Chuck
I have just checked the resistance of the AVC line to ground at C58 and it is 970K ohms--Does this seem about right?

Chuck

No, that's waaaay low. You should read around 2.4Meg. Probably one of the four capacitors I listed is very leaky or even shorted. (assuming that C58 is new and good).

Suggest you replace these caps one at a time, starting from the front end and check the resistance to ground at each step.

If its not one of these caps, there isn't much left to suspect except the tube sockets.

--Chuck

N7RHU--Chuck
Thanks, I'll replace these capacitors early next week. While you mentioned C22 did you mean C32? C22 is a mica trimmer.

Chuck

Yes, that should have been C32, the coupling cap between the crystal filter and the first IF grid. and C33, coupling cap to the RF amp grid. Try them one at a time, until the resistance from AVC bus to ground suddenly jumps to >2Meg.

As a former owner of an SX-99, I would suspect one of the IF transformers as they are the kind that are prone to silver-mica migration disease. Normally, the symptom of this is static. But, depending on where the fault is, it could cause the problem you are having without static.

Dave

N7RHU--Chuck
I replaced the suggested 4 capacitors but the problem with the AVC continues. The resistance to ground at C-58 remains 97K ohms(it was a typo when I said 970K ohms). The receiver continues to work as before. You mentioned the possibility of tube sockets problems--is there a way to check those out?

Thanks for your help on this , Chuck--perhaps the SX-99 is possessed!

Chuck -
Make an extra copy of your schematic and get some colored pencils.
First trace the AGC line in color from R1 to R70, to C58, to R12, and to the AVC switch. Connect your ohmmeter from the top of C58 to ground.
In these instructions "top"" bottom" refers to location on the schematic diagram...not physical position on the chassis. As you perform the following steps, mark your progress in a different color.

Now, first disconnect the bottom of R1 and record the ohmmeter reading, then reconnect it. Next disconnect the bottom of R70 and again record the ohmmeter reading, ...and reconnect. Next, lift the lead between the top of C58 and pin 2 on T2, record the ohmmeter reading, and reconnect. Next disconnect the lead from the left end of the AVC switch, read the ohmmeter, and reconnect. Finally disconnect the bottom of R12, read the ohmmeter, and also before reconnecting R12 use the ohmmeter to measure its resistance.

At some point in the above steps, the resistance from AVC line to ground should have jumped above 2Meg. That would show us which branch of the circuit has the excess leakage. Then we can go on to find the exact source of the problem.

<edit - right now I would bet on either a defective AVC switch or a wiring error>

--Chuck

N7RHU--Chuck
Thanks for the detailed analysis procedure.

N7RHU--Chuck
In running through the trouble shooting procedures you outlined I found that when disconnecting the AVC bus(and C-58) from T-2 pin 2, the AVC bus measured 2.6Meg. However, pin 2 to ground was 100k which is about the measurement I've been getting at that point all along.
In looking at the schematic it shows that pin 1 of T-2 is connected to R29, a 100K resistor, though Gimmick C-38. And Gimmick C-38 measures less than 1 ohm resistance, measured end to end. While I had intended to work another problem with the BFO circuit later it now seems connected to the problem at hand. The problem is that when AVC is on and the AM/CW switch is moved to CW a very loud howl is produced, but when switching AVC off, the BFO works normally.

Does the C-38 Gimmick look like the source of the AVC problem based on these new measurements?

Chuck

Chuck,

If you're saying the gimmick is shorted, then that's definately a problem. There should be no electrical connectivity between the two halves of the gimmick. That's just two pieces of wire twisted to give you a few pF to lightly couple the BFO output into that 2nd IF stage. So yes, if your gimmick is shorted and putting the 100k in parallel with the grid line, it could easily be loading down the AVC.

arnie

I agree. Its easy to check - just untwist and separate the two wires of the gimmic. If your AVC returns to normal, a short in the gimmic was the problem.

Remember, a gimmic is formed of two pieces of INSULATED wire.

--Chuck N7RHU

N7RHU-- Chuck
There doesn't appear to be a gimmick--what is in the receiver is a single conductor shielded wire with the shield cover braiding grounded. Perhaps a previous owner removed the gimmick. Do you know what length of two insulated wires should be twisted together to equal the 2pF that's listed for the C-38 gimmick?