I was chasing down distortion I was getting on strong stations, and it seems that a half megohm resistor running from the AVC line to a -1.8VDC bias seemed to be the cause. Weak signals sound fine with or without this resistor, and strong stations became less distorted without this resistor. Radio seems quite sensitive, so I don't think it is a weak converter or IF tube.

The engineers at Tasma (or their OEM) surely had a good reson for having this resistor, part #18 on the schematic in this link but the radio works better without it. Consumer products don't have unnecessary parts in them, drives up costs and kills profits. I may have some other defective component that I'm covering up? Or maybe Australian cities didn't have that many strong stations like New York City does (about six 50,000 watt flamethrowers)?

Wa2ise,

The resident gurus must be away or busy .... I haven’t seen this radio before and certainly no guru ... but some ‘thoughts’ for you. I assume that you are shorting out resistor 18 which improves the strong signal quality (??).... which suggests that the bias on one or both grids fed by 18 (Cap connections on the Mixer and Det-AVC-IF amp) is more appropriate. I also assume that voltage at junction of 32 and 33 is -1.8V.

I would measure both bias voltages when 18 shorted and use that as my reference voltage. As per the Table, this bias should be Eg approx -1.8V.

There should be little voltage drop across 18. Measure this voltage with weak and strong signals. Measure bias at both Cap connections with strong and weak signals. Disconnect AVC feed resistor 17 and repeat measurements. You are looking for a change of either bias voltage beyond their ‘comfort’ zone, hence causing clipping of the signal. Check capacitor 2, as a fault here could shift the bias of the Mixer. Similarly, any current leakage across 3,4 or 14.

Be very aware that this set is back biased via R32 & R33 the -1.8 Volts will only be developed if the correct circuit current is flowing.

You may also get Hum if the first filter cap (30) is connected to the chassis and not the CT, which is the most negative point. -6V above the chassis, which is positive with respect to CT.

One hopes there is not a 6V6 in the output.

Have the paper & electrolytic capacitors been replaced? Not much point continuing if they haven't as they can contribute significantly to that issue as can wrong & out of spec resistors.

Marc

Thanks all, the set has been recapped, paper and electroytics, and I disconnected resistor 18 (left one end dangling so I could put it back later). Now I can see the usual AVC action that I would see in an All American 5ive tube set, which is quite similar to this set as far as the AVC is concerned. The 2meg resistor 17 measures correctly, as do all the other resistors in this set. No 6V6, it has the EL33 as per the diagram.

I think resistor 18 may have been a substitute for having a cathode resistor on the IF tube (which in an AA5 set doesn't have the detector diodes). But resistor 18 loads down the AVC line on strong stations, cutting its effect, like a leaky AVC filter cap would. For the time being, the radio works fine with one end of resistor 18 disconnected, but later on I'll revisit this, and later I can reconnect it if I want to.

There are a couple of things to look at. The bias will tell you if the current draw is correct. Should it not be there is a danger that one of the tubes is getting tired.

The resistors in the voltage divider from the plates are also part of the AGC. Normally we do not have a lot of issues with Silve Mica caps here, but there is always one, normally on B+, that will, on the odd occassion , let go. I would revisit them, the divider resistors. and the IF's note these are not 455kHz (458) and that can impinge on the AGC volts being developed.

I would actually like to see a 0.047 cap on B+ after the field to decouple RF, especially if the filters were paper and have been replaced with Electrolytic's.

I am not sure on your comment re cathode resistors?

None of these tubes have self bias as it is being developed by the back bias. That saves resistors. The more curent flowing throught the backbias the more is the bias applied as the chassis is positve with respect to the CT.

This is why I enquired about 6V6. It needs -12.6 V and will draw different current .... that will upset the bias applied to all tubes as will damaged back bias resistors. They often cook if there is a heater cathode short. Which is a good reason to not "beef" them up.

This is an occassion where a clean Sig gen and a, rarely used for diagnostics, CRO may give some clues.

Take heart, I have a hacked set on the bench and it has serious issues, due to the technological genius of the moron that did it.

I have plenty of scope for AGC testing I have clear line of sight to 2 Transmitters within a 11Km radius & with 66 feet of antenna feeding to the bench, there is never an issue finding signal.

Marc

Tried replacing the last cap I hadn't yet touched in theAVC circuit, #23, the detector filter, 260pF, with a silver mica but it made no difference. At this point, I'll leave it be till later, when either myself or a future collector/caretaker can find what I have missed. That's why I didn't remove the half megohm resistor that used to connect to the -1.8VDC bias, but just left one end in the air (I did cap it off with a small wirenut).

It's possible that one or more tubes have a problem, but I don't have any EBF2 or EL33 spares. The EL33 does have almost the right cathode current, as deduced from the DC voltage I measure (5.5VDC, should be 6V) I see at the power transformer centertap. And the set does behave like my other tube AM radios (strong signal and weak signal reception works pretty well), and also my experience that tubes are usuall not the problem in old sets.

The most likely SM's to let go are those with HV DC on them

EBF 2 was superceded with EBF32/ EBF35

Marc