I recently got an Ampeg J12T. Ever since I got it I thoughit sounded way too distorted. I first thought i hada blown speaker, but Other speakers are also way distorted. I'm nit sure if it's distorted by design or if the amp has a problem.
It has an el84 push pull output with a cathode bias. The bias resistor is 120 ohms, and the schemo says a voltage across the cathode resistor of 10.4 volts. That means the current is 43mA. With a plate voltage of 300 volts that sounds way too hot to what I see for a recommended idle current, more like 28mA iswhere it should be.
I have ordered some other resistors, and about 200 ohmswill give me around 28mA according to my psice sim for the el84.
Isn't 43mA way too high for an EL84 at 300V? Wouldn't that drive it into distortion with almost no headroom?
Dennis
It has an el84 push pull output with a cathode bias. The bias resistor is 120 ohms, and the schemo says a voltage across the cathode resistor of 10.4 volts. That means the current is 43mA. With a plate voltage of 300 volts that sounds way too hot to what I see for a recommended idle current, more like 28mA iswhere it should be.
I have ordered some other resistors, and about 200 ohmswill give me around 28mA according to my psice sim for the el84.
Isn't 43mA way too high for an EL84 at 300V? Wouldn't that drive it into distortion with almost no headroom?
Dennis
The 43mA is for both tubes, so if the are perfectly matched that's 26.5mA each tube. Also remember cathode current includes screen current too.
Not too high for class A.
If it is distorted, I doubt it is a bias issue. Look at the phase inverter, are both sides putting out signal? Is it clean? Use a scope or signal tracer to view the waveform through the amp, where does the distortion come in.?
If it is distorted, I doubt it is a bias issue. Look at the phase inverter, are both sides putting out signal? Is it clean? Use a scope or signal tracer to view the waveform through the amp, where does the distortion come in.?
Open resistor issue.
I measured the bias currents of the EL84 tubes. First one was right at 28mA, which is a good number.
The second one was 0mA. Hmmm? Seems a tad low?
Found R23 open, the plate grid resistor, I think that's what it's called anyway.
So all this time I was listening to a one tube push amp, not a push pull. Or maybe it was a pull?
I looked at the output with a scope, and it looked fairly normal? It did seem to be slightly more one sided. But I guess everything is AC coupled so it's not so obvious. I am surprised t didn't sound worse than it did?
Will update when parts get here.
I measured the bias currents of the EL84 tubes. First one was right at 28mA, which is a good number.
The second one was 0mA. Hmmm? Seems a tad low?
Found R23 open, the plate grid resistor, I think that's what it's called anyway.
So all this time I was listening to a one tube push amp, not a push pull. Or maybe it was a pull?
I looked at the output with a scope, and it looked fairly normal? It did seem to be slightly more one sided. But I guess everything is AC coupled so it's not so obvious. I am surprised t didn't sound worse than it did?
Will update when parts get here.
Puzzled?
I found a 100 ohm resistor and put that in. But now my idle current at the cathode is 14mA, which is really cold for an el84. I guess it makes sense that the bias current is 1/2 of what is was when just one tube was working.
The amp sounds a whole lot better.
But isn't 14mA way too low for a cathode biased class AB amp with 320V B+ ?
I tend to like my tubes on the cool side, so I may just quit messing with it, but I just wonder if it was biased a bit higher what it would sound like?
It sounds a whole lot better now that both tubes are working.
I found a 100 ohm resistor and put that in. But now my idle current at the cathode is 14mA, which is really cold for an el84. I guess it makes sense that the bias current is 1/2 of what is was when just one tube was working.
The amp sounds a whole lot better.
But isn't 14mA way too low for a cathode biased class AB amp with 320V B+ ?
I tend to like my tubes on the cool side, so I may just quit messing with it, but I just wonder if it was biased a bit higher what it would sound like?
It sounds a whole lot better now that both tubes are working.
It really should run around 80mA both cathodes together. 10+V at 120 Ohms.
That puts EL84 very near its rated Pdiss.
While fixed-bias amps can be run "cool", cathode bias amps pretty much have to idle enough current to cover their load impedance at the supply voltage they get. Otherwise they stumble going from small sounds to LOUD and back. I've never seen a happy EL84 amp run cool. (The usual problem is they burn the tolex off.)
Conversely, if a happy cathode-bias amp loses a tube, it can "play" astonishingly "well". The remaining tube picks up half the current the dead tube isn't taking. It is a rich hot SE amp. The poor OT is lopsided with DC current, but many PP OT designes allow for some, and too-much may be acceptable for guitar. The remaining tube cooks its guts out, but sometimes they hang in for many many hours.
As you have not changed the OT or the supply voltage, I would assume Ampeg's engineers got it pretty-near right. Or at least more-right than where you appear to be now, dead tube or 14mA.
That puts EL84 very near its rated Pdiss.
While fixed-bias amps can be run "cool", cathode bias amps pretty much have to idle enough current to cover their load impedance at the supply voltage they get. Otherwise they stumble going from small sounds to LOUD and back. I've never seen a happy EL84 amp run cool. (The usual problem is they burn the tolex off.)
Conversely, if a happy cathode-bias amp loses a tube, it can "play" astonishingly "well". The remaining tube picks up half the current the dead tube isn't taking. It is a rich hot SE amp. The poor OT is lopsided with DC current, but many PP OT designes allow for some, and too-much may be acceptable for guitar. The remaining tube cooks its guts out, but sometimes they hang in for many many hours.
As you have not changed the OT or the supply voltage, I would assume Ampeg's engineers got it pretty-near right. Or at least more-right than where you appear to be now, dead tube or 14mA.
So?
What you're saying is it's still broken?
The working tube did show signs of getting too hot. It has Tube Store logo painted on and that was darkened.
I guess I need to check the cathode resistor. 120 ohms and 10.4v is what it says on the schematic.
Is it possible I just have a set of tubes that run really cold in this circuit. That is another variable, the actual gain of the tube.
I will check the cathode resistance and voltage next.
What you're saying is it's still broken?
The working tube did show signs of getting too hot. It has Tube Store logo painted on and that was darkened.
I guess I need to check the cathode resistor. 120 ohms and 10.4v is what it says on the schematic.
Is it possible I just have a set of tubes that run really cold in this circuit. That is another variable, the actual gain of the tube.
I will check the cathode resistance and voltage next.
I was telling someone about the amp and how it appeared the working tube had gotten very hot. It occured to me that since the 28mA was only flowing in one tube there was no current in the opposite direction to cancel the flux in the output transformer. So when the transformer saw an increase in current from a signal being present it would drive the core closer to saturation and the inductance of the go way down, resulting in a huge spike of current.
The amp is sounding great.
I was still struggling with measuring the actual bias current. I have a bias checker that goes in series with the tube, and runs the cathode out to a couple banana plugs. I put those in the mA input on my dmm.
I measured 14mA using a small cheap dmm I have. But I measured the voltage across the cathode resistor to be like 7v.
I checked the cathode resistor and it is 120 ohms as it should be.
I tried another dmm and this time I measured 25mA.
So I think what I am seeing is the effect of the dmm's internal series resistance. The one that measured 25mA is about a $40 dmm, while the one that measured 14mA is about $5, so that makes sense, the cheaper one would have a higher resistance.
So basically what I am discovering is in a cathode biased amp I can't use a cheap ammeter with a lot of internal resistance. The el84 is also part of the problem, since its bias voktage of 10v is so low. A couple extra volts and the current is affected by a lot. If the bias voltage is -25 volts lije on my 6v6 amp, another volt or three isn't as big a deal.
I think for Christmas I am going to ask Santa for a nice Dmm.
I was still struggling with measuring the actual bias current. I have a bias checker that goes in series with the tube, and runs the cathode out to a couple banana plugs. I put those in the mA input on my dmm.
I measured 14mA using a small cheap dmm I have. But I measured the voltage across the cathode resistor to be like 7v.
I checked the cathode resistor and it is 120 ohms as it should be.
I tried another dmm and this time I measured 25mA.
So I think what I am seeing is the effect of the dmm's internal series resistance. The one that measured 25mA is about a $40 dmm, while the one that measured 14mA is about $5, so that makes sense, the cheaper one would have a higher resistance.
So basically what I am discovering is in a cathode biased amp I can't use a cheap ammeter with a lot of internal resistance. The el84 is also part of the problem, since its bias voktage of 10v is so low. A couple extra volts and the current is affected by a lot. If the bias voltage is -25 volts lije on my 6v6 amp, another volt or three isn't as big a deal.
I think for Christmas I am going to ask Santa for a nice Dmm.
You got two meters? Put the "Internal resistance" one on current, then measure its resistance with the other meter set to ohms. Rather than resistance, I might think the general accuracy of the meter might be suspect. How many current ranges does it have? Let us be brave and think they claim a 1% accuracy. If it has one 10A scale, 1% of 10A is 100ma. hard to get an accurate measure of 20ma when the accuracy of the meter is within 100ma. Maybe you have a 1A range. 1% of 1A is 10ma. So your meter reading can be 10ma either side of the true current and be within the accuracy of the meter.
Am I correct in assuming your adaptor does not have a 1 ohm series resistor, but instead is a complete break for inserting a current meter? That fits your description. Most adaptors have a series 1 ohm resistor you measure VOLTAGE across. 1 ohm translates 1ma into 1mv.
You have a cathode resistor that measures 120 ohms? Use Ohm's Law then. Eliminate the adaptor. 7v across 120 ohms makes 58ma flowing. That would be 29ma per tube if there is a pair. 50ma (25ma x 2) through 120 ohms makes 6v. All that is close to the 25ma meter. yes?
Meters have accuracy and precision. Accuracy means if you measure a 10v standard voltage, the meter will read 10v. The accuracy of the meter allows a bit of variation. Precision means , well, a precise reading. Your precise meter will read 10.089v every time. They are not the same. A precise but inaccurate meter might read 11.03v every time on the 10v source. An accurate but imprecise meter might read 9.87v one time and 10.02v another time on the 10v standard source.
Am I correct in assuming your adaptor does not have a 1 ohm series resistor, but instead is a complete break for inserting a current meter? That fits your description. Most adaptors have a series 1 ohm resistor you measure VOLTAGE across. 1 ohm translates 1ma into 1mv.
You have a cathode resistor that measures 120 ohms? Use Ohm's Law then. Eliminate the adaptor. 7v across 120 ohms makes 58ma flowing. That would be 29ma per tube if there is a pair. 50ma (25ma x 2) through 120 ohms makes 6v. All that is close to the 25ma meter. yes?
Meters have accuracy and precision. Accuracy means if you measure a 10v standard voltage, the meter will read 10v. The accuracy of the meter allows a bit of variation. Precision means , well, a precise reading. Your precise meter will read 10.089v every time. They are not the same. A precise but inaccurate meter might read 11.03v every time on the 10v source. An accurate but imprecise meter might read 9.87v one time and 10.02v another time on the 10v standard source.
Yep
At this point I agree, just measure the voltage.
It is sitting right at 14W per tube at the moment which makes it 100% for the 7189 tubes from The Tube Store.
That seems high to me, but it sounds really nice, so I guess I should leave well enough alone.
I might try a slightly higher value resistor, just to lower it a bit and extend the tube life. Or maybe wait to see if tube life is an issue?
At this point I agree, just measure the voltage.
It is sitting right at 14W per tube at the moment which makes it 100% for the 7189 tubes from The Tube Store.
That seems high to me, but it sounds really nice, so I guess I should leave well enough alone.
I might try a slightly higher value resistor, just to lower it a bit and extend the tube life. Or maybe wait to see if tube life is an issue?
> So? What you're saying is it's still broken?
Well, something sure is, as you later discovered. (If you can't trust your meter, you shouldn't measure electricity; I admit that this specific failing is very odd.)
http://www.diyaudio.com/forums/instruments-amps/292328-chinese-iron.html#post4900763
> But this isn't a class A amplifier.
Cathode-bias amps *have* to run very close to class A. If they had the large excursions of current implied by class AB/B, the bias would change as you played.
EL84 at 300V can play fine with cathode current, not steady, but rising <20% from idle to full-roar. That's hardly breaking out of class A.
14W is not wrong for modern guitar "12W" EL84. Me, I might increase the cathode resistor 20%-30% to get down near 12W. OTOH EL84 are incredibly low-price, and don't seem to fail much even when abused.
Well, something sure is, as you later discovered. (If you can't trust your meter, you shouldn't measure electricity; I admit that this specific failing is very odd.)
http://www.diyaudio.com/forums/instruments-amps/292328-chinese-iron.html#post4900763
> But this isn't a class A amplifier.
Cathode-bias amps *have* to run very close to class A. If they had the large excursions of current implied by class AB/B, the bias would change as you played.
EL84 at 300V can play fine with cathode current, not steady, but rising <20% from idle to full-roar. That's hardly breaking out of class A.
14W is not wrong for modern guitar "12W" EL84. Me, I might increase the cathode resistor 20%-30% to get down near 12W. OTOH EL84 are incredibly low-price, and don't seem to fail much even when abused.
Last edited:
- Status
- Not open for further replies.