Hey folks - first post here and just wondering if anyone could offer some direction. I am building a low power tube guitar amp without following any particular design. I'm working on the output stage which is single ended, class A with one EL84.
For what its worth, the previous stages before the output stage operate correctly. Also, I do get dirty little bursts of guitar output from the speaker if I strum the strings and/or have the volume pot up enough.
My schematic is shown below with measured voltage values and I'm looking to have the following observations explained:
1. There doesn't seem to be any current flowing through the EL84. I know the middle grid voltage should be a bit lower than B+ but with a 4.7k resistor I'm only dropping 0.1V across it. Bigger resistor I guess?
2. My grid voltage fluctuates like crazy when I probe it with the multimeter (between -50V and -11V??).
3. I'm using a Hammond 125CSE OT using the 10k primary and 16 ohm speaker. However when connected into the circuit, I measure 194 ohm primary and 1.6 ohm speaker!!
4. During the initial heat-up process, I get a large AC burst out of the speaker which quickly decreases in amplitude and frequency down to nothing.
5. I know the EL84 ONLY uses 6.3Vac for the filaments but I started building by using 12.6Vac per my 12AX7 tubes. But I notice once the standy switch is closed and main power delivered, the heater voltage of the EL84 actually drops itself down to 6.3Vac automatically!!
Thanks for any help in advance.
For what its worth, the previous stages before the output stage operate correctly. Also, I do get dirty little bursts of guitar output from the speaker if I strum the strings and/or have the volume pot up enough.
My schematic is shown below with measured voltage values and I'm looking to have the following observations explained:
1. There doesn't seem to be any current flowing through the EL84. I know the middle grid voltage should be a bit lower than B+ but with a 4.7k resistor I'm only dropping 0.1V across it. Bigger resistor I guess?
2. My grid voltage fluctuates like crazy when I probe it with the multimeter (between -50V and -11V??).
3. I'm using a Hammond 125CSE OT using the 10k primary and 16 ohm speaker. However when connected into the circuit, I measure 194 ohm primary and 1.6 ohm speaker!!
4. During the initial heat-up process, I get a large AC burst out of the speaker which quickly decreases in amplitude and frequency down to nothing.
5. I know the EL84 ONLY uses 6.3Vac for the filaments but I started building by using 12.6Vac per my 12AX7 tubes. But I notice once the standy switch is closed and main power delivered, the heater voltage of the EL84 actually drops itself down to 6.3Vac automatically!!
Thanks for any help in advance.
Well, I chose my 470 ohm resistor to set cathode bias at -10V which should give me ~22mA current through the tube, correct?
I get that the tube is obviously in cut-off because the grid voltage is so negative relative to the cathode. But I can't explain why it's so negative, or why it's so unstable.
I get that the tube is obviously in cut-off because the grid voltage is so negative relative to the cathode. But I can't explain why it's so negative, or why it's so unstable.
the control grid of your el84 is floating ...
add 470k resistor from el84 control grid to ground (connect between 0.1uF and 5.6k)
add 470k resistor from el84 control grid to ground (connect between 0.1uF and 5.6k)
1. No current because, as others have pointed out, there is no resistor that references the EL84 grid to ground. 200K to 470K should work.
2. See above.
3. If you're just measuring the transformer, that's the DC resistance of the copper coils. The AC impedance of the speaker reflects back to the primary of the transformer with the proper impedance.
4. Magnetic coupling maybe? Maybe the current through the EL84 is fluctuating before it settles down into cutoff.
5. As the heater warms up, resistance increases. If you overvolt it, it will overheat and resistance will increase quickly. Increased resistance will let less current flow, so it won't go into thermal runaway, but it will use ALOT more power. It should stabilize at somewhere above 6.3v. Your power transformer could be undersized and the voltage could be dropping for all the heaters, which would be bad for the input tubes. I'd recommend putting a 8 ohm 10 watt resistor in series with the EL84 heater. Otherwise your EL84 lifespan could be thrillingly short.
BTW, you're running pentode mode with no feedback? That should be...interesting. If it were me, I would ditch the 4.7K resistor between G3 and B+ and go with a 100 ohm between G3 and the tube anode to run in triode mode. Less power, but gives the tube greater internal feedback.
2. See above.
3. If you're just measuring the transformer, that's the DC resistance of the copper coils. The AC impedance of the speaker reflects back to the primary of the transformer with the proper impedance.
4. Magnetic coupling maybe? Maybe the current through the EL84 is fluctuating before it settles down into cutoff.
5. As the heater warms up, resistance increases. If you overvolt it, it will overheat and resistance will increase quickly. Increased resistance will let less current flow, so it won't go into thermal runaway, but it will use ALOT more power. It should stabilize at somewhere above 6.3v. Your power transformer could be undersized and the voltage could be dropping for all the heaters, which would be bad for the input tubes. I'd recommend putting a 8 ohm 10 watt resistor in series with the EL84 heater. Otherwise your EL84 lifespan could be thrillingly short.
BTW, you're running pentode mode with no feedback? That should be...interesting. If it were me, I would ditch the 4.7K resistor between G3 and B+ and go with a 100 ohm between G3 and the tube anode to run in triode mode. Less power, but gives the tube greater internal feedback.
Does the EL84 filament still light? 12 volts is enough to toast it pretty much immediately, so no current draw. Before you do anything else, find yourself some 6.3 volts, which can also be used for the 12AX7 (wire pins 4 and 5 together as one side and use pin 9 as the other side) and add a grid leak resistor. Should make you a nice guitar amp.
All good fortune,
Chris
All good fortune,
Chris
Per forum policy this really belongs in the Instruments and Amps forum, see the headers at the top of tubes/valves and I&A for an explanation of why.[/moderation off] 😀
As has been mentioned you really need that 470K grid resistor, once that has been addressed you will have plate current, and likely the input transformer will fry shortly thereafter.. You either need a proper power transformer available at Edcor at a reasonable cost or at least a separate 6.3V transformer for the 6BQ5 filament - the fact that it drags the voltage down to 6.3V in combination with the switcher is a good indication that the transformer is massively overloaded.
A traditional transformer based supply will provide more of what you are looking for in terms of guitar amplifier performance..
Here is one possibility which if used in conjunction with something like a 6CA4 rectifier tube would probably be pretty decent. Use duncan amps psud2 to design the input filter: http://www.edcorusa.com/p/604/xpwr013_120
(Note that there is also a 240V version as you provided no location information. Please provide as that will help people to suggest appropriate parts and vendors.)
and http://www.duncanamps.com/psud2/index.html
The volume control at the input is redundant as it's in series with your guitar's volume control. If you move it to between the 2 triode sections of the input tube and use a 500k or 1M pot, you'll end up with a typical guitar amp gain/volume control.
..todd
..todd
Yup.
Also, guitar pickups are high impedance. I think the vol pot on a typical Gibson guitar is 500k. That 10k pot on the input of the amp would load down any normal electric guitar pickup. It would probably sound dull and flat, lifeless.
I hope that helps.
Also, guitar pickups are high impedance. I think the vol pot on a typical Gibson guitar is 500k. That 10k pot on the input of the amp would load down any normal electric guitar pickup. It would probably sound dull and flat, lifeless.
I hope that helps.
what he saidalso I would change the 470k of first gain stage to 1M if you have passive pickups
Thanks everyone - I got a chance to tinker tonight and unfortunately still having stability issues.
I updated the following based on your feedback (image below):
1. Added grid leak resistor (tried connecting to the control grid as well as between the 0.1u cap and 5.6k, and also tried values of 470k and 300k)
2. Put stage in triode mode by connecting 100 ohms from the middle grid to the anode.
With these changes, voltages still fluctuate like crazy (including control grid voltage - this measures anywhere from -10V to -24V and the cathode which fluctuates from +5.5V to +13V), though I'm at least getting current through the tube (though that fluctuates too - anywhere from 11mA to 37mA is what I measured).
Now, tube current/conduction changes with its input signal, correct? I'm noticing I get a continuous low frequency thump out of the speaker after the tubes warm up...maybe the amp is just very noisy for some reason and this is the cause of the bias fluctuations? Just a thought.
And this is the PT I'm using - I have a hard time believing I'm overloading it if it can really source up to 4 amps. I also have a 3 amp fuse which has never blown.
24V, 4A Center Tapped (12-0-12) Transformer-MPJA, Inc.
Thanks again in advance for any help!
I updated the following based on your feedback (image below):
1. Added grid leak resistor (tried connecting to the control grid as well as between the 0.1u cap and 5.6k, and also tried values of 470k and 300k)
2. Put stage in triode mode by connecting 100 ohms from the middle grid to the anode.
With these changes, voltages still fluctuate like crazy (including control grid voltage - this measures anywhere from -10V to -24V and the cathode which fluctuates from +5.5V to +13V), though I'm at least getting current through the tube (though that fluctuates too - anywhere from 11mA to 37mA is what I measured).
Now, tube current/conduction changes with its input signal, correct? I'm noticing I get a continuous low frequency thump out of the speaker after the tubes warm up...maybe the amp is just very noisy for some reason and this is the cause of the bias fluctuations? Just a thought.
And this is the PT I'm using - I have a hard time believing I'm overloading it if it can really source up to 4 amps. I also have a 3 amp fuse which has never blown.
24V, 4A Center Tapped (12-0-12) Transformer-MPJA, Inc.
Thanks again in advance for any help!
To obtain triode mode you would directly connect the plate and the screen. If you want it switchable take a three pole switch and on the center lug connect to screen, the plate gets connected to one side of the switch, and the screen supply would be on the other side of the switch.
Where are you getting your B+? How is it filtered? I have a hard time reading the power supply schematic.
Where are you getting your B+? How is it filtered? I have a hard time reading the power supply schematic.
C2 would be your screen supply and would go on one side of the switch, the other side of the switch to the plate, and the center of switch would go the screen of EL84 in series with a 100 ohm resistor.
Also I would change the cathode resistor to 250 ohms otherwise you will have a cold bias especially with a large signal the valve self bias' itself colder. Reading anywhere from 10 to 12 volts dc across 250 ohm cathode resistor is in the ballpark.
If you don't have a center tapped secondary on power transformer you will have to use 4 diodes in a bridge rectification configuration.
Also I would change the cathode resistor to 250 ohms otherwise you will have a cold bias especially with a large signal the valve self bias' itself colder. Reading anywhere from 10 to 12 volts dc across 250 ohm cathode resistor is in the ballpark.
If you don't have a center tapped secondary on power transformer you will have to use 4 diodes in a bridge rectification configuration.
Attachments
Last edited:
So the way I have the 100 ohm wired in my last post is correct for triode mode "only". I'm not yet at the point where I care to make the modes switchable, I'd just like "a mode" to work first.
Below is how I am generating my B+. It's a cheap alternative because the PT I'm using is 4A but very cheap and I feed 16Vdc into this module to get 200Vdc (45mA current, but it's enough).
I have it filtered exactly as shown in my first image post with the capacitors n such. I assume I need to try adding more filtering to ensure my power rail is absolutely stable - that might keep my bias voltages more stable as well?
Below is how I am generating my B+. It's a cheap alternative because the PT I'm using is 4A but very cheap and I feed 16Vdc into this module to get 200Vdc (45mA current, but it's enough).
I have it filtered exactly as shown in my first image post with the capacitors n such. I assume I need to try adding more filtering to ensure my power rail is absolutely stable - that might keep my bias voltages more stable as well?
I don't think that device will give you the results you are looking for. Also if you haven't fixed your heater configuration for the EL84 than you are asking for some headaches.
There is a reason good iron is costly 🙂
There is a reason good iron is costly 🙂
Last edited:
Mount the 470k to the other side of the 5.6k to have the 5.6k act as a stopper, which is it's purpose. Leave the grounded side of the 470k grounded.
I moved this resistor and tried many values ranging from 100k up to the max 1Meg.
Now I am having an issue with the speaker motorboating. I posted a video here for those interested to hear the actual sound (it's kind of funny - and sorry my phone didn't turn the view sideways) --> DIY tube amp motorboating - YouTube
When I play guitar through the amp, it's like I'm fighting the motorboating to get it out of the speaker. I have to strum hard enough or it's like the guitar signal is eaten up by the motorboating.
As I was troubleshooting, I noticed that if I touch my Fluke 115 multimeter to the plate of the very first Preamp tube (12ax7), the motorboating "goes away"; that is, it lessens in amplitude greatly but also seems to rise in frequency greatly. The noise is much less noticeable, but the speaker is still humming pretty good. I didn't capture this part in my video, but any explanations for this behavior? What is my meter doing to the circuit that would help stability?
My power supply voltage is very steady but I've added more filtering caps without any noticeable benefit.
From what I've read, motorboating could be caused by the grid leak resistors being too high which causes current to not be returned to ground quickly enough as it comes out of the tube...but all my values are the standard 470k, etc...
Decoupling cap is not decoupling properly due to high impedance in preamp is my guess. Do you have an esr meter?
- Status
- Not open for further replies.