All this makes perfect sense, and I get it. However, what has me stumped is that the voltage at the oscillator tube socket heater pins measures 6.3 V while the voltage at terminal 11 measures around 12 V with the oscillator tube removed. I would expect the voltage to be the same at both points whether the oscillator tube is in or not, right?. So here are my findings summarized:

With oscillator tube removed:
Voltage between terminal 11 and chassis ground: 11.98 V
Voltage between pins 3 & 4 of osc. tube socket: 6.28 V

With oscillator tube in:
Voltage at terminal 11 and chassis ground: 11.95 V
Voltage between pins 3 & 4 of osc. tube socket: 4.5 V

How would you interpret these results?

Don

What you have is the equivalent of two globes in series hooked to say a 12v battery, negative to chassis.

The ballast "B" is one globe the heater (H) is the other. Point eleven is the midpoint terminal between them.

when both globes B & H are burning, theoretically (if they are the same) Half the battery voltage appears at point eleven (1:1 voltage divider). However: If the globe H blows, there is no current flow. Therefore there is no loading so and all the globes go out. The point eleven will then go to the voltage of the battery (as will the heater pin in this case with no tube).

The Ballast (Barretter)is a special sort of globe (Iron filament in Hydrogen atmosphere) it is the current controlling element. So what I see is that circuit operating somewhat correctly but with low volts.

To restore the volts, you adjust V3 to a higher resistance and if it does not make it, there is a possibility that the Selenium rectifier is getting tired, or there is a cap drawing current as well. Electrolytic caps draw current, as do leaking other ones.

Try adjustment of r3 with the tube in & see where that all goes.

Note I will be out of utility power from 22:30 UTC for abt 6 hrs.

Marc

My point exactly. I expect the heater pin voltage to be the same as that at point 11 with no tube, but in my case it is not and I have not been able to figure out why.

In any case, I was able to bypass the HV part of the PSU by using my Heathkit HV regulated power supply. By applying 145 V to point 8 (heater supply still coming from the transformer via the selenium rectifier), I was able to confirm that the RF and IF oscillators are working (I was able to get readings on a frequency counter connected to the RF and IF outputs and the readings were almost spot on with the frequency dial readings on all bands). I was also able to confirm that the audio oscillator section is working. So, despite the discrepancy in the voltage readings, the signal generator appears to be working, and at this point I am quite happy!

Don

The answer is looking at you....... R3 the voltage adjusting pot..... that is a load parallel with the ballast.

Marc

But I was not measuring voltages before and after R3, Both measurements were taken AFTER R3. What am I missing?

Don

If there are any paper capacitors in this piece of equipment do replace them before doing anything else.

Let us ignore the decorations like capacitors & inductors and look at the basics. These are the principal components having a bearing on the voltage. The ballast tube is intended to maintain a constant voltage by heating & cooling in response to fluctuating input voltage due to constant changes on the loading on the utility supply.
This as how I see the circuit at it's most basic. See if this gives a clearer picture of what is going on under normal circumstances. Then with the removal of critical elements of the circuit.

Marc

Marc,
Nice job with the diagrams. I understand perfectly what you are saying. The problem is that my voltage readings are as shown in the diagram. Points X and the heater should be at the same potential since both are after R3, right?



I just got a new 6X4 tube, plugged it in and so far no flashover or other problems.

Don

That would reverrt to the last drawing: I have serious issues with the way that that potentiometer R3 is. If it is installed the way it is described on the circuit, which is crazy. It is eventually going to be a short circuit, controlled by the ballast, which may then fail.

If R3 has failed and the tube, is out you get 12V across the heater pins and you may get quite high volts with it in. If the ballast tube has failed shorted, you will get no regulation. I would expect it to glow like a dull globe if it's working & has some load on it. With no loading I would also expect more than 12V from the rectifier.

Check those parts& report back.

Marc

Check the resistance of each choke that is connected in series between point 11 and the tube heater. According to the schematic you should get 6.3 Vdc at point 11.

For the interested parties the 6AF4 / EC94 heater current is quoted as 0.225 Amps. The resistance of the entire line between 11 and the heater pin (tubed removed, would be interesting). Is there any idea of what wattage the rheostat R3 is. I would wonder if it should be wired as a pot & the tube current draw off of the wiper? In reality if the chokes have resistance and are not wholly reliant on R3 to pull the voltage down I would expect a maximum of around 25 ohm. But without seeing how much wire is on them, in this circuit I would expect less

Read R2 of the voltage divider in drawing as R3 and R1 the ballast. If it proves to be a PIB, we can be naughty & replace the ballast with a fixed resistor and R3 with a Zener.

What we have at the moment is a quasi voltage divider circuit, that will be a little confusing to those who have never dealt with a voltage, divider. Voltage dividers are quite common on screen grids.

Measure the dc voltage at each choke and see where the voltage drops. When you do find the voltage drop check the solder joint before and immediately after the component causing the voltage drop. If all is good measure the resistance of the choke. I am guessing these are RF chokes and should have minimal resistance.

For really good regulation and less load on the transformer, you could use a 7806 regulator, here's a diagram
http://www.circuitstoday.com/6-volt-reg ... using-7806
Don

Actually, it is not reverting to the last diagram, and that's what I find so confusing. It is actually a combination of the last drawing and the other two drawings (with the osc. tube in, 12 V at point 11 and 4.5 V at heater pins and with the osc. tube out 12 V at poiint 11 and 6.3 volts at the heater pins.



I would love to be able to do that, but it is easier said than done. The oscillator section is built like Fort Knox. It has two cylindrical metallic cans surrounding the oscillator circuitry. Many of the chokes are within the narrow space between the two cans and are impossible to get to for measuring resistance or voltages without major disassembly which I don't want to take on at the moment. And you are right, they are all RF chokes to prevent any RF getting into or out of the oscillator section. And yes, the schematic shows 6.3 VDC at point 11, but I am getting around 12V. The oscillator tube heater seems to be getting the proper voltage though, because the signal generator works and I was able to use it as an FM transmitter to transmit music, and as already mentioned, when I measured the voltage at the heater pin it was 6.3 V with the tube removed.

since you cannot easily get to the chokes remove the tube then measure the resistance between point 11 and the heater pin of the tube socket. Should see very low resistance. If not you may have a problem.

Mystery solved!!

Measured the resistance between point 11 and pin 4 (the non-grounded heater pin) of the RF oscillator tube and found it to be almost 1K (964 ohms to be exact). On further inspection I noticed that the wire going to the first RF choke between point 11 and the heater pin 4 had been replaced (looked different than the other wires). I measured the resistance from point 11 to the point of connection of this wire and found that the wire accounted for almost all of the resistance. What was even more curious was that the original wire had been disconnected at this point but was left in place. Initially I thought that maybe the original wire was open and that was why it was replaced by the more modern-appearing wire. But, on checking resistance between point 11 and the disconnected end of the original wire, I found that the wire was good and showed a very low resistance of about 1 ohm. Also, on further inspection I noticed that the ballast tube is a 7TF4 and not a 6TF4 that is specified in the schematic. The ballast tube filament shows continuity. I looked on the web for specifications of 7TF4 vis-a-vis the 6TF4, but was not able to find any information on the differences between the two tubes. This clears up the mystery of the voltage drop between points 11 and pin 4 of the oscillator tube, but it is not clear to me why the instrument was modified in the first place? I am postulating that this modification was made presumably to make a 7TF4 work in place of a 6TF4. But why would someone go through all this trouble when a 6TF4 is probably just as easily available as a 7TF4?

6TF4 here http://www.tubedepot.com/nos-6tf4.html

Maybe it was done to work with what the previous owner had on hand.

Hi

First number is current. 6TF4 would be 600 ma Last number voltage, 4 volts. A 7TF4 is made for 700 ma but also 4 volts. Both were standard Amperite numbers.

Edit - Thought this was interesting, Amperite is asking $105 for 6TF4. 7TF4 is listed at $149.

Thanks for the information, Norm. Wow! $105 and $149 for ballast tubes?? Incredible!

Thanks to the suggestion from Tube Radio, the mystery is solved. I had previously measured the resistance from the start of the series of the RF chokes to the heater pin and since the resistance was quite low, I was puzzled as to why there was a voltage drop. It never occurred to me to check the resistance from point 11 to the heater pin of the RF oscillator tube. Just goes to show that one should never assume anything - my mistake was to assume that the wire between point 11 and the first of the RF chokes was an ordinary wire and would not contribute to any voltage drop when in fact it turned out to be some kind of coated resistance wire which for all the world looked like an ordinary piece of hook up wire.

Many thanks also to Marcc for his many contributions to this thread and his detailed explanations and diagrams. They helped a lot to clarify many things for me.

Don