So I paralleled up my other PCL805 on the class A amp, and it's oscillating at 100Hz like crazy, so much that it triggers the OCP on my PSU!
Any clues on why it's doing this?
The schematic is in the forum.
Any clues on why it's doing this?
The schematic is in the forum.
I found out that the noise occurs when the 2nd tube's anode connects to the other. This still happens when the screen is floating.
Are you talking about the schematic in post #33? If so, there are a number of things wrong with it.
1. There's no bias on the cathode of the triode. Use a 1K resistor (optionally bypassed with a capacitor of 100uF+) or even a red or green LED.
2. The anode resistor on the triode is probably too big, 50K would be better to put the tube in a more linear region.
3. The resistor on the cathode of the Pentode needs to be much bigger - 330 ohms would probably be a good value.
4. There's no grid stopper resistor on the G1 or G2. A few hundred ohms soldered directly on the socket is fine.
4. There's no feedback, so the pentode has a very high output impedance. Quickest fix is to turn it into a triode by connecting G2 to the anode. Less power, but less complexity too. Otherwise you will have to make a feedback loop from the transformer to the input triode.
1. There's no bias on the cathode of the triode. Use a 1K resistor (optionally bypassed with a capacitor of 100uF+) or even a red or green LED.
2. The anode resistor on the triode is probably too big, 50K would be better to put the tube in a more linear region.
3. The resistor on the cathode of the Pentode needs to be much bigger - 330 ohms would probably be a good value.
4. There's no grid stopper resistor on the G1 or G2. A few hundred ohms soldered directly on the socket is fine.
4. There's no feedback, so the pentode has a very high output impedance. Quickest fix is to turn it into a triode by connecting G2 to the anode. Less power, but less complexity too. Otherwise you will have to make a feedback loop from the transformer to the input triode.
Technically it's not a Baxandall tonestack, it's a James, which is a derivative of the Baxandall designed for guitar use. It works just fine for guitar and the caps don't need to be different values if you use a linear rather than a log pot. The plots don't come out symmetrical, but, meh, it's guitar. It also takes into account the input and output impedance.
Besides, everything to the left of R7 on the schematic is only for guitar use.
Hey
Alright, I will do them and test again.
BTW, the screen grids are connected to the anode. I was just trying to see why the amp oscillated when I added the second tube's anode.
Thanks!
You probably find the course at the input side: lost connection to a grid leak resistor or a colt joint at gound level. Seen these pages?
Well, nothing's disconnected or anything.
That page seems to have a lot of info on 805s. And no, I had not seen it before. Thanks for the link!
Results:
1. Adding a cathode resistor to the triode just quietens the amp. I added a blue LED and it helped with the frequency response.
2. Lowering the value to 50kOhms also decreases the output amplitude.
3. I increased the 96Ohms to 500 because it was the only high wattage resistor pair I had and it reduces the bass, but helped with the linearity.
4. I added a 5.1k on the triode grid and it eliminated the squeaking, but I still have a lot of hum without anything connected which I think is because there are many transformers and chokes near the tube?
Also, connecting the screen grid to the anode improves the frequency response, but decreases the power.
Oh BTW, I don't have a socket. I'm using those dupont (I think they're called) breadboard wires.
Thanks for the suggestions!
Last edited:
1. Blue LED may bias the tube too far the other way. The aim is for about 1-2v across the cathode LED or resistor for the triode.
2. The triode operating point is wrong. You need to fix #1 and then #2 will fix itself.
3. Add a largish value capacitor across the 500 ohm resistor, at least 100uf, but 470-1000uf would be better. Gain and bass should improve.
4. The hum is probably coming from the power supply. You may need better filtering or to move/rotate the transformers apart from each other.
Connecting the screen to grid effectively turns the pentode into a triode. Less power but much more linearlity and driving ability, especially if the pentode is operated with no feedback like yours.
Last edited:
Hey
I added a 220Ohm resistor to the grid of the pentode and the screen is attached to the anode.
I will test these new ones today and tell you.
Thanks again
1. I replaced the blue LED on the triode with a green LED, I think it eliminated a bit of the distortion.
2. From the LED's light, I can see that the triode is oscillating. The 220Ohm on the pentode has prevented the pentode from oscillation.
3. I added a 820uF across the resistors and it not only fixed the low bass problem, but it also increased the volume.
4. The 2 transformers are next to each other and are interfering with each other which I think is not a problem since they are the same transformer and do the exact same job from the same power source. The choke was rotated 90° from the power transformers. I think the hum is present because of my caps. I'll need to change them.
And finally, connecting the screen to the grid basically shuts the pentode down. However, connecting it to the B+ increases the output power, but adds a lot of distortion. The best solution was to connect it directly to the anode.
The overall power has decreased, but it sounds MUCH MUCH clearer and better. Also, now that I look at the old pics of the tube in the previous configuration, it was red-plating on a tiny spot. But now it doesn't have any red-plating.
Questions: is there a way to increase the power with the same tube without getting more distortion? Will it help if I add a small amplifier on the tube amplifier input? To increase the max 2V P-P to something more like 5V or something.
Thanks man. You've helped a lot!
Unfortunately, Single Ended class A is going to low efficiency, maybe 10-20% max. With a dissipation rating of about 10 watts, that means about 1-2 watts max. And that's with everything perfect, which this isn't. Going push-pull could quadruple your power. Increasing the input probably won't help - it would just put one of the tubes into clipping.
I noticed there are no high voltage capacitors on the power supply before and after the choke - those would help immensely. I'm thinking that the choke isn't high enough value to affect the 100Hz ripple coming from the 120v section of the power supply. Going from a L to a CLC would improve things, and maybe even another RC for the triode section.
I noticed there are no high voltage capacitors on the power supply before and after the choke - those would help immensely. I'm thinking that the choke isn't high enough value to affect the 100Hz ripple coming from the 120v section of the power supply. Going from a L to a CLC would improve things, and maybe even another RC for the triode section.
I understand. I will get an EL34 if it's available. The seller hasn't responded if it's available or not, and if it is, it'll be less than $2. I might get a bit more (A lot more LOL) power with an EL34.
I couldn't find any higher value choke than this. And yes, there are no capacitors after the choke. I will add a 330uF. If my oscilloscope hadn't broken, I would've told you the results. I know that the choke reduced the ripple by a factor of 6, back when I had the scope.
Last edited:
Update: OK, I found out that the power is basically the same as the V1.0 which sounded trash. I thought it was lower because at max volume, it'd give like half the power. Now the problem was on my phone. The equalizer basically reduces the overall volume to be able to give the "effects". But there's a downside to this V4 or whatever, and it's that it clips very obviously when the input voltage is higher than a certain value. It "clicks" when it clips.
Any ideas on what's causing this click?
Update 2: Adding a 330uF capacitor after the choke drastically improved the high frequencies and I can't hear any hum.
BTW, the voltage across the green LED is about 2.38V. Should I change it out with a red LED?
BTW, the voltage across the green LED is about 2.38V. Should I change it out with a red LED?
For that low of B+ (170v), you probably want 1-1.5v between cathode and ground. A red led might get you lower, down to maybe 1.9v depending on the LED. Better, but not ideal. A resistor would probably get you there too (500 ohms seems about right), you may need to bypass it with at least a 100uf capacitor if you need a little more gain. Another option would be to use two silicon diodes in series, each dropping 0.7v. Make sure they're set the correct way, i.e. not blocking the current. You're merely exploiting their P-N junction voltage drop to give you the voltage offset you need.
With a 330uF cap after the choke and the 2 diodes in series with a 470uF cap across the diodes, the gain (If present) is unnoticeable and the weird clipping issue still remains 🙁
The voltage drop across the diodes is 1.11V BTW.
I'm gonna try adding a grid stopper resistor on the triode. 5.1k probably and see if it can help with the clipping.
I'm gonna try adding a grid stopper resistor on the triode. 5.1k probably and see if it can help with the clipping.