Okay, thanks for explaining that! That makes sense.
Yep. I feel it's a shame that the Hammond got scrapped too (all I had was the amplifier in awful shape), but the organ it came out of was stored VERY poorly and was damaged by mice in many places. No working or good restoration candidate organs were harmed.
BJosephs, thanks for confirming that for me. I did find a datasheet with the anode characteristics for triode mode, and will probably test both configurations to see which I like the sound of better.
Ti83, I've seen ultralinear circuits, and as much as I would love to build one, I already had the parts and I didn't want to put much into a project that is borderline just a learning experience!
Thanks!
Last edited:
Hi fi suggests you need negative feedback.
Without feedback there will be some distortion.
Look up ultra linear output transformers too.
Without feedback there will be some distortion.
Look up ultra linear output transformers too.
Negative feedback (global) suggests one need a decent oscilloscope.
IlikeTech, do you have one? If not, I would recommend avoiding NFB designs.
I have a Rigol DS1054Z, or an old 20 MHz analog scope my dad has.
The Rigol is mine, works great!
The Rigol is mine, works great!
Why? Even with a 'scope in the room, I usually just stuck in a resistor and listened.
If you think you need 37dB NFB for hyper-high damp factor then yes a beginner will want a way to see the squeal. But damping factors of 5 down to 1 (<14dB) take the curse out of pentodes driving "conventional" speakers and there's two typical results: "tamer" or "rude/squeal" (NFB reversed).
Yeah, 5:1 NFB on a junk transformer may have supersonic peaking. But the church-organ transformers were not junk. And his source is a cellphone!! Which may honestly be better than some of the cassettes I have used. But lily-gilding is not needed.
That'll do (both of them)
Just because you never had cancer doesn't mean millions of people don't die from it every year, you know...
Well I have one, so I'm not sure it matters either way. I would be just as willing to do it sonically as visually on a scope screen!
I looked at some of the designs mentioned earlier, and they all used two triodes in a different kind of phase inverter. Is using 1/2 of the AX7 as a gain stage and then using the other half as a cathodyne phase inverter okay?
I looked at some of the designs mentioned earlier, and they all used two triodes in a different kind of phase inverter. Is using 1/2 of the AX7 as a gain stage and then using the other half as a cathodyne phase inverter okay?
Also, I can use 1/4 watt resistors as grid leaks and things where the wattage and voltage ratings aren't exceeded, correct?
Well, it's taken a while, but I have an update. My dad made the aluminum chassis plate for everything to be mounted on, and I've gotten the circuit built, with some modifications. I had the tubes wired as shown in my schematic, in true tetrode mode, and they really didn't sound great. The cymbals and highs were kind of harsh and hard to listen to. I did try adding feedback and that helped, but it seemed to create a 50khz oscillation, which seemed abnormal, and it was still harsh sounding. I did fix some distortion by changing from a 5Y3 rectifier to a 5U4, as the voltage is higher and it can supply more current without too much voltage sag.
In the end, I went triode wired mode, with the screens connected to the anode by 100 ohm resistors. I will play around with a little bit of feedback tomorrow, as it does still sound a touch distorted, but it sounds much better in triode mode than tetrode mode. How much less power do I get from wiring them as triodes?
Thanks! I'll post some pictures soon!
In the end, I went triode wired mode, with the screens connected to the anode by 100 ohm resistors. I will play around with a little bit of feedback tomorrow, as it does still sound a touch distorted, but it sounds much better in triode mode than tetrode mode. How much less power do I get from wiring them as triodes?
Thanks! I'll post some pictures soon!
One or both of your old 6v6 tubes may be weak, you need to measure the current that goes trough each tube to be sure that it is the same on both, and that the value is correct. The easiest way to do so is to use a separate cathode resistor on each tube, and measure the voltage from each cathode to ground when the volume control is turned to zero. Classic automatic bias circuits circuits such as yours use a common cathode resistor and capacitor for both tubes because in the past the capacitor was a bulky and expensive part. Today is best to use separate cathode resistors, they must have twice the value of the single common resistor. A 2W resistor is good enough and you may use a series or parallel combination of two 1W resistors; I use a 7W one on my builds to avoid instant destruction of the resistor if the tube shorts internally. Leave the cathode capacitors disconnected and check how the sound will change, the gain will be lower but the distortion will be reduced too. The 6V6 pinout on your schematic is wrong, cathode is pin 8.
Hi, thanks for replying. I'm using a later schematic that Ketje posted, and it does in fact have seperate cathode resistors already. 680 Ohm 2 watts each. I think the schematic is wrong because I modified a KiCad symbol because I couldn't find one for a 6V6. I haven't been looking at pin numbers on it, so no worries.
Attached is my current schematic, except for the fact that the feedback loop is currently unconnected and needs to be changed for triode wiring.
How much difference from design voltage is too much? My OT drops about 10 volts on a 6V6, but I'm running about 10 volts high on the plate and screens (Pentode wiring). Is that going to be a problem? Also, I have to use a 1K dropper resistor for the first stage. Is that a problem?
Thanks!
Thanks!
IlikeTech,
I think the input cathode circuit is quite unusual.
Seems quite creative (*).
Try this, from the top:
Cathode
1.6k resistor
100 Ohm resistor
Ground
And, 100 uF across the 1.6k resistor
Then the 1.2k feedback resistor from the transformer secondary connects to the junction of the 1.6k resistor and the 100 Ohm resistor (and the minus lead of the 100 uF capacitor.
Does that seem a little more normal?
The difference is the negative feedback at very low frequencies; but perhaps you want the feedback to be reduced at very low frequencies to prevent output transformer saturation (*).
I think the input cathode circuit is quite unusual.
Seems quite creative (*).
Try this, from the top:
Cathode
1.6k resistor
100 Ohm resistor
Ground
And, 100 uF across the 1.6k resistor
Then the 1.2k feedback resistor from the transformer secondary connects to the junction of the 1.6k resistor and the 100 Ohm resistor (and the minus lead of the 100 uF capacitor.
Does that seem a little more normal?
The difference is the negative feedback at very low frequencies; but perhaps you want the feedback to be reduced at very low frequencies to prevent output transformer saturation (*).
Last edited:
I'll give that a shot! Can't take credit for that NFB loop though, Ketje added that in an earlier post.
At this point, my circuit is running about 350 on the tube plates, 285 on the screens. I did recalculate the bias resistor for the power tubes, and they are currently not bypassed with capacitors. I'll probably upload my current schematic here again soon.
At this point, my circuit is running about 350 on the tube plates, 285 on the screens. I did recalculate the bias resistor for the power tubes, and they are currently not bypassed with capacitors. I'll probably upload my current schematic here again soon.
Changing the cathode circuit as 6A3 propose makes not much diffeence.The feedback voltage reaches the cathode via C2 // R2 not only C2.And the biasing cathode resistor goes from 1k6 to 1k7, changing the voltages on the tube a little.
Changing the output to triodes to get lower anode impedance (better damping of the speakers) but not decoupling the cathode resistors does the opposite, not very productive.
The voltage on the output tubes is the power supply minus drop on the OPT and minus the cathode voltage.
In most cases ±10% on the voltage doesn't do any harm.Just take care the output tubes don't get to hot.
Try a capacitor bypassing the feedback resistor (R5), start with 1nF (1000pF).Perhaps the harsh sound is from to much overshoot.
Mona
Changing the output to triodes to get lower anode impedance (better damping of the speakers) but not decoupling the cathode resistors does the opposite, not very productive.
The voltage on the output tubes is the power supply minus drop on the OPT and minus the cathode voltage.
In most cases ±10% on the voltage doesn't do any harm.Just take care the output tubes don't get to hot.
Try a capacitor bypassing the feedback resistor (R5), start with 1nF (1000pF).Perhaps the harsh sound is from to much overshoot.
Mona
IlikeTech,
If I had noticed that Mona proposed that input cathode circuit, I would not have even commented on it.
As was said in post # 39, it does not make a lot of difference.
Especially since the ECC83 cathode impedance is very high, versus the capacitive reactance at low frequencies of the bypass cap.
If I had noticed that Mona proposed that input cathode circuit, I would not have even commented on it.
As was said in post # 39, it does not make a lot of difference.
Especially since the ECC83 cathode impedance is very high, versus the capacitive reactance at low frequencies of the bypass cap.