Hi
I'm repairing an old pre-war radio for a friend who knows I play around with valve/tube amplifiers. It has a single ended output stage with a 6g6-g pentode running at 135V / 11mA through an output transformer mounted on the speaker . There is only low sound level even at full volume.
I tested the output transformer out of circuit and got the following results:
-Primary DC resistance 600 ohms, secondary DC resistance about 0.2 ohms
-7.5 V AC@50 Hz from a small "wall-wart" applied to the primary gives 95 mV at the unloaded secondary, drops to 88 mV with a 4 ohm load. This implies a turns ratio of about 80, or impedance ratio of 6400:1. the 6g6-g requires a load of 12500, but these figures give a load of 22400 with a 3.5 ohm speaker.
-If I apply the 7.5V to the secondary (with primary unloaded) it appears to short out the supply with not much voltage measured anywhere
-While applying the 7.5V to the primary, if I short out the secondary leads the supply voltage drops by only 0.01V, i.e. it is not loading the supply.
Is this transformer dead? I would have thought that I would get no output at all if this was the case, or are there circumstances where you still get some sound even with a short somewhere?
Note - the radio as supplied had a long length of flex soldered onto the speaker which was still connected to the secondary, so it appears someone in the past had been driving the speaker directly from another source, maybe becuse they liked the sound of the old console radio. Unfortunately this had put strain on the speaker leader wire which eventually broke, and the voice coil is now open circuit.
I'm repairing an old pre-war radio for a friend who knows I play around with valve/tube amplifiers. It has a single ended output stage with a 6g6-g pentode running at 135V / 11mA through an output transformer mounted on the speaker . There is only low sound level even at full volume.
I tested the output transformer out of circuit and got the following results:
-Primary DC resistance 600 ohms, secondary DC resistance about 0.2 ohms
-7.5 V AC@50 Hz from a small "wall-wart" applied to the primary gives 95 mV at the unloaded secondary, drops to 88 mV with a 4 ohm load. This implies a turns ratio of about 80, or impedance ratio of 6400:1. the 6g6-g requires a load of 12500, but these figures give a load of 22400 with a 3.5 ohm speaker.
-If I apply the 7.5V to the secondary (with primary unloaded) it appears to short out the supply with not much voltage measured anywhere
-While applying the 7.5V to the primary, if I short out the secondary leads the supply voltage drops by only 0.01V, i.e. it is not loading the supply.
Is this transformer dead? I would have thought that I would get no output at all if this was the case, or are there circumstances where you still get some sound even with a short somewhere?
Note - the radio as supplied had a long length of flex soldered onto the speaker which was still connected to the secondary, so it appears someone in the past had been driving the speaker directly from another source, maybe becuse they liked the sound of the old console radio. Unfortunately this had put strain on the speaker leader wire which eventually broke, and the voice coil is now open circuit.
Yes. I think it was leftover from an old cordless phone charger, easiest/safest source of low voltage AC I could lay my hands on. It's definitely AC.
Here's the good news: sounds like your transformer is fine. The secondary will not have sufficient inductance to look like much beyond a short to your wall wart. Also, the impedance you measured could be different at midband- many radios used transformers with severely limited bass range.
This wouldn't explain the loss of volume. If secondary was indeed shorted (or at least shunted by extremely low R) it would explain the low volume regardless of the volume setting.
@tikiroo: have you confirmed that everything is operating as it should up until the grid of the output tube ? Use AC voltage source (your improvised supply + potentiometer could work in a pinch) and AC voltmeter to confirm that you're getting enough signal excursion at the grid of the output tube. If you don't, trace the signal backwards through the circuit in order to locate the fault (and remember you're dealing with high voltages !).
@tikiroo: have you confirmed that everything is operating as it should up until the grid of the output tube ? Use AC voltage source (your improvised supply + potentiometer could work in a pinch) and AC voltmeter to confirm that you're getting enough signal excursion at the grid of the output tube. If you don't, trace the signal backwards through the circuit in order to locate the fault (and remember you're dealing with high voltages !).
The voltages around the 6G6 appear OK, 5.8V on the cathode across a 510 ohm resistor, 135V on the plate, 140 on the screen - pretty close to the RCA tube data. But touching an unearthed probe to the grid doesn't give the expected "buzz" (Yes I'm careful about the high voltages present). Grid to earth is 470k.
I'll try connecting a known good signal directly to the grid and see what I hear.
As far as the turns ratio of the transformer goes, I had thought that 50 Hz might be a bit low to be measuring performance. I'll have to borrow a signal generator and try again at 1 kHz.
Thanks for the advice!
Edit - All coupling capacitors are new and good - the old ones were wax paper/oil 70 years old and all leaky.
I'll try connecting a known good signal directly to the grid and see what I hear.
As far as the turns ratio of the transformer goes, I had thought that 50 Hz might be a bit low to be measuring performance. I'll have to borrow a signal generator and try again at 1 kHz.
Thanks for the advice!
Edit - All coupling capacitors are new and good - the old ones were wax paper/oil 70 years old and all leaky.
Last edited:
Um, it's also worth noting that 11 mA and 135V isn't going to be loud anyway - you're looking at 0.75W *at most* (realistically more along the lines of 0.5W).
Screen voltage is slightly higher than Plate voltage. Is this normal for the radio?
Is there a capacitor between the Cathode of the 6G6 and ground?
Is there a capacitor between the Cathode of the 6G6 and ground?
Update - when I measured the primary resistance this morning it was up to 850 ohms. When powered up there was a reasonable volume for a while then it went very quiet again. Measuring the DC resistance again it was now only 450 ohms, and jumped around when I tapped the outside of the transformer. So it seems there is an intermittent short in the primary.
I'll have to track down a replacement. Maybe a 100V/8ohm line output transformer would work at a pinch (5W rated using a 0.8W tap) -11mA shouldn't saturate the core. As Arnulf said, output is only rated at 0.6W, hi-fi it ain't!
I'll have to track down a replacement. Maybe a 100V/8ohm line output transformer would work at a pinch (5W rated using a 0.8W tap) -11mA shouldn't saturate the core. As Arnulf said, output is only rated at 0.6W, hi-fi it ain't!
DC resistance changing by a factor of 2 definitely indicates transformer fault 🙁
Line transformers are not designed for SE duty. I have used one in SE arrangement and the sound was definitely acceptable for radio - human ear is extremely imprecise instrument and will tolerate all sorts of signal mutilations. And besides, lack of bass and power gives the sound that "antique" radio feel 🙂
Power transformer with suitable ratio (230V : 6V will get you in the neighbourhood of what you've got there now) and large enough core will work better; try to locate one rated for at least twice the quiescent power wasted (135V * 11 mA * 2 or more W, in your case it's still a rather low figure).
Line transformers are not designed for SE duty. I have used one in SE arrangement and the sound was definitely acceptable for radio - human ear is extremely imprecise instrument and will tolerate all sorts of signal mutilations. And besides, lack of bass and power gives the sound that "antique" radio feel 🙂
Power transformer with suitable ratio (230V : 6V will get you in the neighbourhood of what you've got there now) and large enough core will work better; try to locate one rated for at least twice the quiescent power wasted (135V * 11 mA * 2 or more W, in your case it's still a rather low figure).
- Status
- Not open for further replies.