Put a 260kc signal and scope on the i.f. transformers in the 70. The first i.f. input transformer peaked out at 260kc spot on, the second i.f. peaked out around 175kc - could not peak at 260kc, so I checked another output i.f. transformer coil out of another early 70 - it too peaked out NOT at 260kc, but around 225kc, could not get to peak right even with adding the trimmer cap. I read this in and old Radio Service Shortcuts:
'PHILCO 90 (With one -17 output) Weak, stations do not track with dial settings . . . although the i.f. peak is given as 260 kc. by the manufacturer some were built to align at 175 kc'.
Has anyone else ever run in to this? Was this a factory coil error or service error? Shouldn't it peak at 260 also?
I think you mean '47 output, not 17. Type 17 is a triode similar to the 27 except it has a 14V filament. It was designed for use in Philco model 46 and some other 115 volt DC-only radios.
True, some service info called for a 260 kc IF in the model 90 with single 47 output, but that is in error - both the first (2-45) and second (1-47) versions of the 90 have a 175 kc IF.
Now, if an IF transformer in a 70 won't properly adjust at 260 kc, I would suspect the IF transformer itself - did someone pop in an IF from an early or mid 90, or from a 112 (all 175 kc IFs)? I think it would have to be either the IF transformer, or something wrong with the trimmer condensers.
All model 70 sets were designed with an IF of 260 kc. I've serviced dozens of Model 70 sets and never had that issue.
It is possible someone put an IF can from a 90 thinking they were the same.
Thanks Ron, they typed it wrong in the Service publication, I think I have found what I have been experiencing in the unit, can tune it in, but never right. One of the second if's had two extra wires coming out to nowhere from the bottom of the coil (like a gimmick), so I think these must have been replacements. Sheesh...
Put a Merit 262kc output transformer in. Strangely, all the trimmers peaked from nominal like they should. Seems you can check ohms and inductance on a Bside cap checker - everything looks good until you put volts on these old coils, but maybe that's the way it was.
Experimenting with this radio, decided to bias the 47 tube's grid to -16.5 like the normal characteristic spec sheet calls for on the 47, was running around -20+ as is measuring with a Dynascan 177 Vtvm to ground. Took 4 meg ohm to bring it down to -16.5. Produced a most beautiful non-garbled sound especially with music and very pure heavy bass, but as in whack a mole, the hum suddenly increased, extra capacitance did not cure it. Think it is ac heater hum as the spec sheet on the 47 says the input resistance at the grid should not be over 50k - thinking it soaks out the hum - the high resistance leaves it nowhere to go. Philco used 240k to feed the grid with the negative bias, so did they bias the tube wrong? I am researching the input impedance meaning a what it affects. No grid stoppers on this one.
Keep in mind that in 1931, the typical voltmeters used to make the voltage measurements at the factory had an input impedance of 1000 ohms/volt, which loads down the circuit under test and produces false readings. Only they did not know this at the time.
In other words, voltage readings taken under identical conditions today (110 or 115 volts AC input, depending upon published specs) with a DMM or a good VTVM will be higher than Philco's published voltage readings.
I did consider this as always, the 177 vtvm measures the plates at 250 v, my Klein meter reads 269 or so on the plates, so I 'sort of figured' I might in the correct ballpark with the vtvm, but is is 11meg input vs. what they used to measure. I was still a bit taken back at what the 47 bias can do. I still have not decided if I can get a good player out of this radio, if I do, I will strip it down for a full restore from a clean chassis, still not there yet. I might try changing to 47 filament to DC with a bridge rectifier if the transformer allows it. I am not trying to keep it original, even with the 2.5 volt bulbs and their shortcomings.
Changing the resistor connected to the 47 grid from 240K to 4 meg will not change the actual bias, only make the bias unstable. The 47 bias is determined by the radio's DC current drain flowing through the 240 ohm section of the wirewound resistor 37. The voltage drop across this resistor creates the bias for the 47, not the grid resistor.
If you increase the value of the grid resistor, the tube bias is more affected by grid current due to gas in the tube or grid emission when the tube overheats. This can cause a runaway condition as the tube gets hotter and hotter, the current keeps increasing. That's why the tube data sheet specifies a max resistance for the grid resistor. It should be no higher than 240K.
If the grid voltage on the 47 is -20V, it means that the 47 or some other component is drawing excessive current through the 240 ohm bias resistor. Excessive current can create more hum as it loads down the power supply. Could be a leaky coupling capacitor, or a gassy 47 tube. Have you replaced capacitor 31?
Check the voltage across the 240 ohm resistor and see if it is in the neighborhood of 16 to 18 V.
Apologize for not getting back, still playing around with this radio. Yes it has been totally recapped and checked out. I see what you are saying, I did monitor it on the vtvm and saw the instability it created esp. on upper volume, still a fantastic sound that I have never heard from it. Proud to say I see many of the problems mentioned in Rider books when working with this radio - digging down into the books and seeing the same things on the scope when I feed it with the generator. - have learned alot from this unit, great tool and reading how they progressed in the 30's.