Roberts R66

[LyntonP]

Gentlemen first of all can I apologise for my lack of understanding and asking the same question more than once. I am a complete novice at this restoration hobby and it has been many years since I have had to apply ohms law or any other formula to electrical circuits. Similarly I admit that I struggle transposing a circuit diagram to the underside of a chassis or circuit board. Thanks for your understanding.
I have rewired the selenium rectifier as per the trader sheet and the radio is working again. 
If I measure the voltage at the output of the rectifier (junction of R16 and C31 to LT-) it is2,7V.
If I measure the voltage across C29 it is 1.42V.
R16 out of circuit measures 5.0 ohms.
R15 out of circuit measures 6.5 ohms.
Current flow to S7 100mA.
My mains voltage from my isolating transformer is 244 volts.
I appreciate the suggestion when replacing resistors to start high and work lower until a correct value is found but I do not have access to a stock of components and would have to buy from a remote supplier with postage charges higher than the value of a small number of components.
Resistors R15 and R16 are sizeable components is this due to their age? If 100mA flows through a 3.9 ohm resistor would this produce only 0.3W? (Power is IxV watts)?
So would a 1/2 watt resistor be ok?
Even though R15 and R16 are high in value I am tempted to leave them be since the voltage across C29 is less than 1.5v (even now may be a tad high)?
Finally thanks for the ongoing help. Your discussions are invaluable to me.
Finally I have attached two more photos of the wip. The second one clearly shows the Original connection of one side of the rectifier (wire floating in space).
Lynton

[Terry]

Photo of valve

It's a DK96, the markings are the manufacturing code.

Yes, etched into the glass so that they can't rub off like the other markings!

Manufactured in the Valvo factory in Hamburg during the second week of February, 1967! The only thing we don't know about it is the name of the girl who assembled it!

[ppppenguin]

Let's try a simple approach. Use a couple of silicon diodes (1N400x) to substitute for the old MR2. The ends with the band are connected together and to C31 etc. Reconnect the caps and resistors. Conenct a voltmeter to C29 so you are measuring the actual filament voltage. Switch on. If it's between 1.2V and 1.4V then you're OK. No more work needed. If it's over 1.4V switch off. If it's over 1.8V switch off ***** quickly.

[LyntonP]

I was going to try a couple of diodes but was nervous that the voltage would be so high it would destroy the valve filaments. That’s why I asked if anyone knew of a suitable dropper resistor I could put in addition to the two low value resistors already in the circuit.
Lynton

[LyntonP]

And it is V1 which has the unmarked valve not V3 as I originally stated. However it does seem to have the correct valve as other members have discovered.
I do have both the manufacturers and traders sheets for the R66 but make the mistake of trying to read both at the same time. 
No excuse for the valve numbers as they are the same on both sheets.
Lynton

[Terry]

If I measure the voltage across C29 it is 1.42V.
R16 out of circuit measures 5.0 ohms.
R15 out of circuit measures 6.5 ohms.
Current flow to S7 100mA.

Resistors R15 and R16 are sizeable components is this due to their age? If 100mA flows through a 3.9 ohm resistor would this produce only 0.3W? (Power is IxV watts)?
So would a 1/2 watt resistor be ok?

The current flow should be 125mA unless there is a battery economy switch but that should be inoperative for mains use.

This range of valves has 25mA filaments but the DL96 has two filaments! (Actually a 2.8V filament with a centre tap.) The two sections are connected in series in many mains operated receivers (but not this one!) so effectively there are five 1.4V filaments in series. For parallel connection, as here, the two sections are also connected in parallel, so it draws 50mA with the other 3 valves drawing 25mA each, a total of 125mA. It would be worth looking at the connections to the DL96 to check it is connected ok. Also check the valveholder pins from above for damage.

If you can find that missing 25mA, it will bring your excess voltage down!

Those resistors were very common at one time. The size of the resistor is quite small but what you are seeing is the relatively thick ceramic tube that surrounds it.

Power is I x V watts but for the figures you are using, current and resistance, you need to transpose the formula to find V. V = I x R, so the formula now becomes W = I² x R (normally written as I²R)

So 0.125 x 0.125 x 3.9 = 0.061W - so yes ½W would be more than adequate! Even 125mW would be fine! However, it looks at the moment that you need higher values than 3R9.

A thought that crosses my mind is that you measured the voltage at around 1.1V when the set stopped working and those resistor values are high - which could account for it but you are now missing 25mA  somewhere so try this as an experiment: connect the meter across the LT line and place a finger on the top of the DL96 and move it around with a circular motion whilst keeping your eyes on the meter.  If it suddenly drops to 1.1V, you've found your problem - either a faulty of just dirty valve holder or a dry joint under the chassis.

I'm assuming you are using a modern electronic meter and not an old analogue meter which is badly out of calibration?

Find that missing current because something isn't right there, then we can address the subject of replacement resistor values.

Oh - nearly forgot. No need for the apology - we all have to start somewhere and if you'd upset anybody you wouldn't be getting all this wonderful advice, would you?

[pwdrive]

Lynton, re: post#31....power dissipation for a 3.9 Ohm resistor with 100mA current flowing through it is Isquared*R which equals 0.039 Watts (0.1*0.1*3.9)

Lawrence.

[LyntonP]

Let's try a simple approach. Use a couple of silicon diodes (1N400x) to substitute for the old MR2. The ends with the band are connected together and to C31 etc. Reconnect the caps and resistors. Conenct a voltmeter to C29 so you are measuring the actual filament voltage. Switch on. If it's between 1.2V and 1.4V then you're OK. No more work needed. If it's over 1.4V switch off. If it's over 1.8V switch off ***** quickly.

I could put 2 diodes in and bring the mains up slowly on the variac whilst monitoring the LT voltage!
Lynton

[LyntonP]

Not really sure how to quote part of a post but Terry’s advice about the the resistors and missing 25 mA is much appreciated.
Lynton

[ppppenguin]

Didn't realise you had a variac. Excellent idea.

To quote part of a post just quote the whole thing and delete the bits you don't want to quote.