Hi, in addition to hi-fi, I play bass guitar and am into separates, preamps, power amps.
I'm looking at some tube power amps for my bass guitar. There are a few tube power amps out there, Mesa Boogie Strategy 400 Stereo, Peavey Classic 120/120, a couple more. These are typically push/pull designs with either 6L6 or 6550 power tubes. Only the Peavey has a built in feature to combine the channels for one mono output equal to the combined watts of each channel, so I know tube amps can be bridged.
So if I have a two channel power amp, 100 watts a channel, and I want to get one mono output of 200 watts going to one speaker, is that parallel bridging? How difficult or expensive would it be to design bridge circuitry if I buy an amp that is only available as two channel output?
Is it as simple as paralleling the inputs and outputs together as was once recommended for the Dynaco ST70?
I'm looking at some tube power amps for my bass guitar. There are a few tube power amps out there, Mesa Boogie Strategy 400 Stereo, Peavey Classic 120/120, a couple more. These are typically push/pull designs with either 6L6 or 6550 power tubes. Only the Peavey has a built in feature to combine the channels for one mono output equal to the combined watts of each channel, so I know tube amps can be bridged.
So if I have a two channel power amp, 100 watts a channel, and I want to get one mono output of 200 watts going to one speaker, is that parallel bridging? How difficult or expensive would it be to design bridge circuitry if I buy an amp that is only available as two channel output?
Is it as simple as paralleling the inputs and outputs together as was once recommended for the Dynaco ST70?
The Dynaco method SUCKS. Most of the time, you get bad sound and lots of heat. 
The fool proof method is to drive the 2 sections with 180o out of phase signals and connect the voice coil across the "hot" taps. A 16 Ω voice coil connects to 8 Ω taps and an 8 Ω voice coil connects to 4 Ω taps. Unless very unusual O/P "iron", with 2 Ω taps, is present, a 4 Ω voice coil may not be used.
What is happening is that the voltage swings of the 2 sections get summed in the load. Each section "sees" only 1/2 of the load, which is why things get wired up, as described.
The fool proof method is to drive the 2 sections with 180o out of phase signals and connect the voice coil across the "hot" taps. A 16 Ω voice coil connects to 8 Ω taps and an 8 Ω voice coil connects to 4 Ω taps. Unless very unusual O/P "iron", with 2 Ω taps, is present, a 4 Ω voice coil may not be used.
What is happening is that the voltage swings of the 2 sections get summed in the load. Each section "sees" only 1/2 of the load, which is why things get wired up, as described.
Thanks Eli.
I also asked on DIYTube, and a membed offered both methods, but stated the paralell method better for tubes than bridge. Oh well!
I am looking at two channel tube power amps in the 120~200wpc range if that helps. All the designs have massive output transformers, these amps weigh in from 50~75 lbs.
What is the preferred way to take my my mono input signal and make two input signals, one being 180 degrees out of phase? Put two 1:1 transformers on each input, but one has output leads reversed?
I also asked on DIYTube, and a membed offered both methods, but stated the paralell method better for tubes than bridge. Oh well!
I am looking at two channel tube power amps in the 120~200wpc range if that helps. All the designs have massive output transformers, these amps weigh in from 50~75 lbs.
What is the preferred way to take my my mono input signal and make two input signals, one being 180 degrees out of phase? Put two 1:1 transformers on each input, but one has output leads reversed?
Last edited:
Don't forget that transformers have to be driven by low impedance circuitry.
Sowter's model 8920 is quite suitable.
A differential gain block is another way to go from "single ended" to "balanced". The Schmitt implementation might be a good choice, as it does not need a B- rail. RK1 provides bias. Replacing RK2 with a constant current sink (CCS) will force symmetry between the 2 triodes. Use a high gm/low RP type like the ECC99 or 6H30П (6n30p).
Sowter's model 8920 is quite suitable.
A differential gain block is another way to go from "single ended" to "balanced". The Schmitt implementation might be a good choice, as it does not need a B- rail. RK1 provides bias. Replacing RK2 with a constant current sink (CCS) will force symmetry between the 2 triodes. Use a high gm/low RP type like the ECC99 or 6H30П (6n30p).
Output impedence of my bass pre amp is approx 1 k ohms or greater with a 0 dB signal, often referred to as Hi Z.
The Mesa Power amp has input impedence of 75 k ohms.
Where does that put me if I just use the Sowter?
Last edited:
Why don't you just look for a good used Peavey Classic 120, seriously doubt you can build one better yourself...
Remember that tubes are more current limited, so bridging may not be the best method for you.
Re: Bridging / My bass tube preamp also has what is known as a DI out, it is a balanced XLR connector delivering a low impedence output in the 45~600 ohm range. Would that be a better source for the 8920 instead of the pre amps normal 1/4" two conductor output that can run well overall 1k of impedence? Since that low impedence DI out is an XLR balanced connector, it has three pins, not sure how to tie all three into the 8920. Two lines I can figure out.
Last edited:
Here are some results with limited testing. Neither test was run for very long, so no opportunity to witness any overheating or other unusual amplifier behavior.
Parallel Mode- I combined the amp's channels A and B inputs via a Y cord, and my 8 ohm speaker was fed off the channels A and B 16 ohm + and - output leads combined in parallel. Output was robust.
Bridge Mode- One of my preamp's had a balanced XLR out rated at <600 ohms impedance. I used pins 2 and 3 for my 180 degree out of phase channel A and B's hot inputs, and pin 1 for a ground. My 8 ohm speaker floated across channel's A and B's 4 ohm output taps. Output seemed weak, I wounder if my preamps XLR output is heavily buffered down. (My preamp's standard 1/4" output runs from 1~3 volts, between 1K~10K impedance.) I did look at some diagrams of the Jensen and Sowter transformers. With strictly transformer based phase inversion, am I loosing any power of that input signal?
Parallel Mode- I combined the amp's channels A and B inputs via a Y cord, and my 8 ohm speaker was fed off the channels A and B 16 ohm + and - output leads combined in parallel. Output was robust.
Bridge Mode- One of my preamp's had a balanced XLR out rated at <600 ohms impedance. I used pins 2 and 3 for my 180 degree out of phase channel A and B's hot inputs, and pin 1 for a ground. My 8 ohm speaker floated across channel's A and B's 4 ohm output taps. Output seemed weak, I wounder if my preamps XLR output is heavily buffered down. (My preamp's standard 1/4" output runs from 1~3 volts, between 1K~10K impedance.) I did look at some diagrams of the Jensen and Sowter transformers. With strictly transformer based phase inversion, am I loosing any power of that input signal?
Last edited:
In the bridged mode, you are connecting the 8 ohm speaker across the (+) on channel A and the (+) on channel B, which is correct.
To make it work (to allow current to flow around the loop comprising channel A OT secondary, channel B OT secondary and the speaker) you also need there to be a connection between channel A (-) and channel B (-). It is common for one side of the OT to be grounded inside the amp, so this connection may already be there for you.
Apologies, if you already know all about this.
To make it work (to allow current to flow around the loop comprising channel A OT secondary, channel B OT secondary and the speaker) you also need there to be a connection between channel A (-) and channel B (-). It is common for one side of the OT to be grounded inside the amp, so this connection may already be there for you.
Apologies, if you already know all about this.
I'm wondering about the available transformer taps for bridge mode as opposed to parallel mode - I suspect a pair of 4 ohm taps might not drive an 8 ohm speaker that well, but having trouble wrapping my head around this at the moment.
What I can say is that usually tube amp outputs are paralleled for more power output, I've never seen any bridged ever. Only discussed here. (Output current does not scale so higher impedance loads are required, unless suitable taps are present on the amp)
Tube amps have relatively high output impedances and behave somewhere between a current and voltage source unlike solid state amps which generally behave as near ideal voltage sources until you start to approach the limits of their current capability. What this means is tube amplifiers usually share pretty well in parallel provided the amplifier gains are well matched.
Edit: Parallel connection gives you twice the available current at the same maximum voltage as an individual amp, to get twice the power the load impedance needs to be half that of the selected taps. Series connection (bridge) gives you twice the voltage at the same current so in theory double the load impedance of the selected taps should work..
Are you sure both phases of the DI output are actually driven, it is not terribly uncommon to present a single ended output on an XLR - handled correctly you still get the hum cancelling of a balanced input, but you do not get the 6dB additional output of balanced mode, and in this case one channel of that amplifier would be undriven - which would match the symptom you describe. DI outputs intended to be plugged into microphone inputs are heavily attenuated.
What I can say is that usually tube amp outputs are paralleled for more power output, I've never seen any bridged ever. Only discussed here. (Output current does not scale so higher impedance loads are required, unless suitable taps are present on the amp)
Tube amps have relatively high output impedances and behave somewhere between a current and voltage source unlike solid state amps which generally behave as near ideal voltage sources until you start to approach the limits of their current capability. What this means is tube amplifiers usually share pretty well in parallel provided the amplifier gains are well matched.
Edit: Parallel connection gives you twice the available current at the same maximum voltage as an individual amp, to get twice the power the load impedance needs to be half that of the selected taps. Series connection (bridge) gives you twice the voltage at the same current so in theory double the load impedance of the selected taps should work..
Are you sure both phases of the DI output are actually driven, it is not terribly uncommon to present a single ended output on an XLR - handled correctly you still get the hum cancelling of a balanced input, but you do not get the 6dB additional output of balanced mode, and in this case one channel of that amplifier would be undriven - which would match the symptom you describe. DI outputs intended to be plugged into microphone inputs are heavily attenuated.
Yes, in bridge mode you halve the load seen by each output, so for maximum power, an 8 Ohm load goes on the 4 Ohm taps, and a 16 Ohm load goes on the 8 Ohm taps. Think of the center of the load as a virtual ground.
Last edited:
I'll take a look at that, could it be as simple as a jumper between the - terminals of channel A to channel B?
Yes, just make sure that the 0 Ohm terminals are grounded because some, like Audio Research, are not. I thnk the Dynace Stereo 70 manual has a description of the parallel connection.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.