Hello - my son has just completed his first valve amp kit which is a ChiFi using 6N2 and 6P14 with dual 6Z4 rectifier valves. After the usual checks, powered on and it's working! We found that one of the channnels has the phase reversed. Ok, we could just connect the speaker in reverse but I'd rather 'fix' the amp.
The OPTs are wired correctly so it could be that "B" (red) and "P" (green) wires on one OPT are internally swapped on the OPT itself? If we swapped the B and G connections for one channel, what would happen to the NFB for that channel?
There's some mains hum as well, so I'm assuming we may have to play around with grounding and/or shielding.
Any upgrade tips? He's interested in trying different valves already 🙂
The OPTs are wired correctly so it could be that "B" (red) and "P" (green) wires on one OPT are internally swapped on the OPT itself? If we swapped the B and G connections for one channel, what would happen to the NFB for that channel?
There's some mains hum as well, so I'm assuming we may have to play around with grounding and/or shielding.
Any upgrade tips? He's interested in trying different valves already 🙂
Looking at the schematic, if the phase was reversed, the NFB would change to Possitive feedback and you would have a howling noise that is uncontrolable.
If the input signals are not out of phase, it must be the speaker wires are wrong. I would check the inputs by connecting both inputs in parallel that will prove the inputs are correct then swap the speaker wires over as the speakers themselves may be wired incorrectly.
Start from the speakers, (PP3 battery on the speaker leads to confirm polarity and then check your input polarity.
As I mentioned, if the NFB loop is connected and works, it cannot be out of phase. Check by disconnecting the NFB loop 18k resistor and check the gain increases.
It is a single ended output stage and they will be prone to 100/120HZ hum, depending where you are in this small World.
If the input signals are not out of phase, it must be the speaker wires are wrong. I would check the inputs by connecting both inputs in parallel that will prove the inputs are correct then swap the speaker wires over as the speakers themselves may be wired incorrectly.
Start from the speakers, (PP3 battery on the speaker leads to confirm polarity and then check your input polarity.
As I mentioned, if the NFB loop is connected and works, it cannot be out of phase. Check by disconnecting the NFB loop 18k resistor and check the gain increases.
It is a single ended output stage and they will be prone to 100/120HZ hum, depending where you are in this small World.
I wonder why the presumably incorrectly phased channel doesn't oscillate, cause, as you said, NFB turns into PFB?
Regarding hum: Any SE amplifier with a simpe PSU like this one will hum more or less. You can increase the 2nd 220µF capacitor and/or replace both 240R resistors by a choke.
Best regards!
Regarding hum: Any SE amplifier with a simpe PSU like this one will hum more or less. You can increase the 2nd 220µF capacitor and/or replace both 240R resistors by a choke.
Best regards!
What Jon and Kay say.
Like so often with ChiFi some of the voltages in the schematic don't make sense.
The voltage drop over the paralleled 240 Ohm resistors in the power supply is 288 - 260 = 28 V. So the current would be I = V / R = 28 / 120 = 233 mA. That ofcourse is an absurd value (way too high). The cathode current of one 6P14 is only 8 / 240 = 33.3 mA, while the current of one section of the 6N2 can't be much more than 1 mA. So the two channels would draw less than 70 mA without input signal.
The voltage drop over the 150K anode resistor of the 6N2 is 210 - 125 = 85 V. So the current would be I = V / R = 85 / 150K = 0.57 mA. The cathode voltage is 1 V, giving a currrent of I = V / R = 1 / (2K in parallel with 18K) = 1 / 1800 = 0.56 mA, which is ofcourse close enough to 0.57 mA. However, the voltage drop over the 36K resistor is 260 - 210 = 50 V. So the current would be I = V / R = 50 / 36K = 1.4 mA, while it should be close to 2 x the current per channel, so close to 2 x 0.57 = 1.14 mA. The difference of 0.26 mA might be small in absolute terms but it's rather large in relative terms.
Note that the total dissipation, so P(a+g2), of the 6P14 is less than 0.0333 x 252 = 8.4 Watt (less because there will be some voltage drop in the primary of the OPT).
Like so often with ChiFi some of the voltages in the schematic don't make sense.
The voltage drop over the paralleled 240 Ohm resistors in the power supply is 288 - 260 = 28 V. So the current would be I = V / R = 28 / 120 = 233 mA. That ofcourse is an absurd value (way too high). The cathode current of one 6P14 is only 8 / 240 = 33.3 mA, while the current of one section of the 6N2 can't be much more than 1 mA. So the two channels would draw less than 70 mA without input signal.
The voltage drop over the 150K anode resistor of the 6N2 is 210 - 125 = 85 V. So the current would be I = V / R = 85 / 150K = 0.57 mA. The cathode voltage is 1 V, giving a currrent of I = V / R = 1 / (2K in parallel with 18K) = 1 / 1800 = 0.56 mA, which is ofcourse close enough to 0.57 mA. However, the voltage drop over the 36K resistor is 260 - 210 = 50 V. So the current would be I = V / R = 50 / 36K = 1.4 mA, while it should be close to 2 x the current per channel, so close to 2 x 0.57 = 1.14 mA. The difference of 0.26 mA might be small in absolute terms but it's rather large in relative terms.
Note that the total dissipation, so P(a+g2), of the 6P14 is less than 0.0333 x 252 = 8.4 Watt (less because there will be some voltage drop in the primary of the OPT).
This amplifier is a good starting point for learning, electrical safety etc is as you know, paramount.
@ TO: Do we know the OT's actual primary impedance or turns ratio? If it is close to 7 kΩ or a bit more, the given plate current would fit (36 mA is better with 7 kΩ). If it is about 5 kΩ, it might be preferable to decrease the cathode resistor value to about 135 Ω (2 x 270 Ω in parallel), which will increase plate current and dissipation as well as the available output power.
Edit: Schematic says 5 kΩ. So I strongly recommend to replace the cathode resistors. I guess it would also decrease distortions.
Best regards!
Edit: Schematic says 5 kΩ. So I strongly recommend to replace the cathode resistors. I guess it would also decrease distortions.
Best regards!
The schematic is a little misleading (no surprise there!). There are two 6Z4 rectifiers just paralleled up to provide twice the capacity. They are not wired to provide one PSU per channel. Only one is shown, however the 2-off 240R resistors and both capacitors provide the B+ for BOTH channels, being a single connection point to the paralled 6Z4 x2-off. With the B+ draw for two 6P14 that knocks it down to something sensible.
When I have time I'll measure some real life voltages (carefully).
Last edited:
No idea of the OPT primary impedance or turns ratio I'm afraid. I could ask the supplier..... Or maybe I could measure with an LCR meter? I have a basic LCR meter but no idea if that uses AC or DC to establish a reading.
With the schematic saying 5k I could experiment with cathode resistors, but I guess also depends on the OPT rating. Cheers.
When we first checked, it was with a pair of Mission 761i. One channel wasn't working so powered doewn, waiting and swapped 6P14 valves. Lo and behold, dead channel moved with the valce. Changed the valve, we got both channels.
At some stage I got one AR bookshelf speaker so we had one of each. I wonder if one speaker is opposite phased to the other, being different makes? They were both copnnected correctly accoring to binding posts on amp and speaker. One more thing to try.
For hum reducing, I would do this:
1. Check wiring and perhaps redo.
2. Make a ground lift for heaters
3. Add a choke for supply
Try in sequence, all might not be needed...
1. Check wiring and perhaps redo.
2. Make a ground lift for heaters
3. Add a choke for supply
Try in sequence, all might not be needed...
Yes, that PSU is simple and no 'natural' noise cancellation. I might try a gyrator cirsuit with a MOSFET rather than bulky choke - will keep this thread updated.
My calculations are based on what you describe, so with the two channels sharing one power supply.
I'll measure when I get a chance. The amp is working, nothing is smelling or getting hot. No red-plating so maybe not idea but within ratings?
Update... I checked speaker wiring and used both AR speakers - same result.
What I have also noticed, with the source in stereo, if I unplug ONE channel RCA, then both speakers dont play music. It's the same for either channel. So, a little more complicated than I thought. I've checked assembly and all looks according to schematic. Next step (when I get a chance) is to remove the volume pot from the circuit and direct wire each input just in case there's some cross-channel issues.
Any other suggestions appreciated!
What I have also noticed, with the source in stereo, if I unplug ONE channel RCA, then both speakers dont play music. It's the same for either channel. So, a little more complicated than I thought. I've checked assembly and all looks according to schematic. Next step (when I get a chance) is to remove the volume pot from the circuit and direct wire each input just in case there's some cross-channel issues.
Any other suggestions appreciated!
Are the input signal grounds connected? If not, then the amplifier could be amplifying just the difference between the L and R signals (when you have inputs to both RCAs) and the outputs from the two channels will be out of phase. With input to only one RCA you would get no output.
T
That describes the symptoms exactly, and I did check signal grounding, which is present. I'll recheck grounding when I get a chance and report back.