GByxbee

TDPRI Member
I have an amplifier with 2 pairs of 6L6GC tubes coupled to an 8 ohm load via an output transformer. Looks like a Class AB circuit with grounded cathodes. According to the manufacturer's schematic, the max plate voltage can be 500 VDC and the max plate current can be 90ma per pair. If I were to adjust my bias to get to a plate current of 45ma per tube and set the plate voltage to 500 VDC, that calculates out to 90 Watts rms max. The manufacturer claims it to be a 100 watt amp. It would seem to me that since the tubes are rated at 30 watts I could increase my plate current and achieve a 100 watt output, but then I would be violating the schematic's 90 ma per pair spec. Should I just leave it at 75 watts (.038 plate currents X 4, with a plate voltage of 470 VDC) or increase my plate currents and get the 100 watts the amp is advertised as being? It's almost like the amp may have been upgraded to 30 watt 6L6GC's from the 25 watt 6L6's and the schematic wasn't updated. I'm tempted to leave it as it is, currently at 470 VDC plate voltages and 38ma plate currents (about 60%). Any opinions?

Ed Storer

Tele-Afflicted
Silver Supporter
IMO, increasing the bias to get 10% increase in output is a good way to burn up tubes faster without noticeable improvement in output.

Phrygian77

Poster Extraordinaire
Maximum plate dissipation and output power are two totally different things. You need to understand what a class AB loadline is to really understand it.

Last edited:

dan40

Friend of Leo's
As Phrygian77 noted, the 25 and 30 watt ratings are really about the maximum amount of heat that the tubes can dissipate and those figures do not correlate to the actual amount of power that the circuit can produce. Larger tubes have a larger internal plate structure and larger envelope therefore they can dissipate more heat than their smaller 6v6 counterparts.

Paul G.

Friend of Leo's
So, you're chasing a number? Why? Does the amp operate as expected? Then I don't know what it is you want. At different times a Twin Reverb has been "rated" at 88 WRMS and 100 WRMS. At onset of clipping, my measurements are always around 67 WRMS. So, if you want a 100 watt amp, just say it's 100 watts and it's a 100 watt amp. What you're proposing to do will not change volume or headroom in any perceptible way, but will add heat and stress.

andrewRneumann

Friend of Leo's
As Phrygian77 noted, the 25 and 30 watt ratings are really about the maximum amount of heat that the tubes can dissipate and those figures do not correlate to the actual amount of power that the circuit can produce. Larger tubes have a larger internal plate structure and larger envelope therefore they can dissipate more heat than their smaller 6v6 counterparts.

I think they do correlate, but not in the way the OP suggests. For class A1, a beam power tetrode / pentode with optimized loading and bias can be expected to make about 45% of it’s maximum plate dissipation in actual clean audio power. In AB1, this number can be expected to double to about 90%. Just ballpark numbers, obviously depends on tube and age of tube. It would be very hard to get 100W from 4x25W tubes if the *rules* are followed. Definitely possible with 4x30W tubes.

The OP seems to be confusing idle plate dissipation with audio power.

GByxbee

TDPRI Member
Thank you to all five respondents! The amp clips at 75W. The amp is supposed to be 100W. I just figured there's something wrong. I'm a solid state guy so this is foreign to me. I'll leave it as it is.

Timbresmith1

Friend of Leo's
Thank you to all five respondents! The amp clips at 75W. The amp is supposed to be 100W. I just figured there's something wrong. I'm a solid state guy so this is foreign to me. I'll leave it as it is.
Clipping is a huge part of the reason people like tubes for guitar amps

Phrygian77

Poster Extraordinaire
@GByxbee my question is how are you measuring power? My Bassman (AA864 output) puts out approximately 45 watts RMS at 1kHz into an 8 ohm resistive load.

GByxbee

TDPRI Member
Clipping is a huge part of the reason people like tubes for guitar amps
I am concerned that with a 50 mV, 0 VDC signal in (don't remember the frequency, but somewhere in between 400 and 1000hz) I start clipping only half way thru the volume knob. If I continue to max volume, the speakers are getting clipping that could almost be seen as a DC signal. As far as I know continuing to play a long song could endanger the voice coils. Is there some spec somewhere that says how much of the signal at the speaker is acceptable? For instance, if 50% of the signal were lost due to clipping would the speaker be potentially damaged? I would also assume that the higher the frequency, the more potential damage to the voice coils.

GByxbee

TDPRI Member
@GByxbee my question is how are you measuring power? My Bassman (AA864 output) puts out approximately 45 watts RMS at 1kHz into an 8 ohm resistive load.
I don't know about Bassman's, but I'm using two 4 ohm non-inductive resistors in series for the dummy load.

Timbresmith1

Friend of Leo's
I am concerned that with a 50 mV, 0 VDC signal in (don't remember the frequency, but somewhere in between 400 and 1000hz) I start clipping only half way thru the volume knob. If I continue to max volume, the speakers are getting clipping that could almost be seen as a DC signal. As far as I know continuing to play a long song could endanger the voice coils. Is there some spec somewhere that says how much of the signal at the speaker is acceptable? For instance, if 50% of the signal were lost due to clipping would the speaker be potentially damaged? I would also assume that the higher the frequency, the more potential damage to the voice coils.
RMS is the typical way to rate speakers/amplifier output.
RMS volts or power (watts) rated into a load value (typically, impedance for our purposes as speakers are an inductive load).

andrewRneumann

Friend of Leo's
Alright, let’s take a step back here. Multiple questions:
What amp is this?
What is the OT reflected impedance load on the power tubes?
What is the plate and screen voltage on the power tubes? (You mentioned 470V for plates, just wanted to confirm both.)
You mentioned 38mA of idle current. How was that measured?
How did you measure 75W before clipping?

The fact that your idle plate dissipation seems to match your clean output power is more of a coincidence than anything. Increasing your bias current will not gain you another 25W of audio power. How old are these tubes? Have you tried a new set?

Phrygian77

Poster Extraordinaire
I am concerned that with a 50 mV, 0 VDC signal in (don't remember the frequency, but somewhere in between 400 and 1000hz) I start clipping only half way thru the volume knob. If I continue to max volume, the speakers are getting clipping that could almost be seen as a DC signal. As far as I know continuing to play a long song could endanger the voice coils. Is there some spec somewhere that says how much of the signal at the speaker is acceptable? For instance, if 50% of the signal were lost due to clipping would the speaker be potentially damaged? I would also assume that the higher the frequency, the more potential damage to the voice coils.

The preamp probably has enough gain to cause the output to clip with the volume at 50% or less. You haven't said what amp this is.

At full clipping, the RMS power and peak power are roughly equal, which is about double what the clean unclipped RMS power is. So, if you wanted to be completely safe, you want speakers with a continuous power rating that is equal to the peak output power. Although, in reality, it's completely unnecessary for a guitar amp.

schmee

Telefied
My guess is it's just part of the common fraudulent selling scheme of businesses. "If the amp makes 90 watts, round up and call it 100 watts."
If the boat is 26.2 feet including the bow rail, 25 on deck, call it 27 feet etc
My Music Man 100 watt amp was verified at 90 watts after being tuned up and checked by the tech.

peteb

Poster Extraordinaire
The amp clips at 75W. The amp is supposed to be 100W. I just figured there's something wrong.

you may get more clean power out with a cooler bias than a hot bias. It actually is supposed to work that way with class AB. Cool it all the way down to class B for maximum clean output power.

although I am using a guitar I measured 50 watts of clean power from a pair of 6L6 tubes.

450 VDC, 32 mA for a bias of around 50%.

bebopbrain

Tele-Holic
The 90ma per pair number is allowed idle current. This doesn't tell us anything about the peak current or maximum power. As @peteb said, run cool to achieve max power; hot idling wastes power without creating music.

B+ = 500VDC. Since this is push pull you can have an AC signal at the plates that is a clean sine wave that is ideally 1000Vac peak to peak or 353Vac RMS.

We can calculate the ideal current to hit 100W:

100W RMS / 353Vac RMS = 0.283A Iac RMS
This is a peak current of about 400ma.

You need more current than this. The output transformer and impedance mismatches drop power. Tubes themselves drop power since they are never perfectly on or off.

GByxbee

TDPRI Member
RMS is the typical way to rate speakers/amplifier output.
RMS volts or power (watts) rated into a load value (typically, impedance for our purposes as speakers are an inductive load).
When I bought these resistors for the dummy load I read somewhere they should be non-inductive.

GByxbee

TDPRI Member
Alright, let’s take a step back here. Multiple questions:
What amp is this?
What is the OT reflected impedance load on the power tubes?
What is the plate and screen voltage on the power tubes? (You mentioned 470V for plates, just wanted to confirm both.)
You mentioned 38mA of idle current. How was that measured?
How did you measure 75W before clipping?

The fact that your idle plate dissipation seems to match your clean output power is more of a coincidence than anything. Increasing your bias current will not gain you another 25W of audio power. How old are these tubes? Have you tried a new set?
This is a Teisco Checkmate 100 amp. Notice "100"! On the schematic, it says "2.8K / 8 ohms" just above the OT. Looks like 400 VDC at the grids. I measured the idle current with a Fluke meter and a 4 test sockets to balance them easier. I measured the 75 W with an O-scope and calculated the power. These are the original tubes. I don't want to invest a large amount of money into it. I just want to sell it off but want to feel good about selling it.

GByxbee

TDPRI Member
The 90ma per pair number is allowed idle current. This doesn't tell us anything about the peak current or maximum power. As @peteb said, run cool to achieve max power; hot idling wastes power without creating music.

B+ = 500VDC. Since this is push pull you can have an AC signal at the plates that is a clean sine wave that is ideally 1000Vac peak to peak or 353Vac RMS.

We can calculate the ideal current to hit 100W:

100W RMS / 353Vac RMS = 0.283A Iac RMS
This is a peak current of about 400ma.

You need more current than this. The output transformer and impedance mismatches drop power. Tubes themselves drop power since they are never perfectly on or off.
Thanks for your time but I feel like I'm going to settle for what I have at 75W after reading the responses I've gotten (including yours).

Replies
15
Views
541
Replies
47
Views
988