2 seperate (switchable) power stages in same amp?

Status
Not open for further replies.
Jump to Latest
I

Illinest

Member
Joined 2014
#1
Problem:
I want a 15-20 watt-ish amp, but i need a 1/4 to 1 watt amp.

(I will rarely have an opportunity to crank it.)

Rather than a master volume or an attentuator i was wondering if it would work out alright to simply design a single ended amp with 2 separate (redundant) tube power stages.

I could use a single 12ax7 as the preamp stage and then use a 3-way switch to select either 1 watt or 20 watt operation.
Would require an additional tube and an additional output transformer - correct?
Would require some careful switching to make sure that each power stage stays out of the circuit when not in use - correct?

Is there any precedent for this? Any designs i could look at? Read about?

Is there any reason why it might not work?
 
#4
I'm afraid i don't understand the suggestion.

I'm pretty new to this - just starting to make my first amp - and i'm not sure what the series/parallel switch would do aside from reduce voltages.

I thought that tubes had to have specific voltages and currents or else they wouldn't run properly or be damaged. I'm trying to guess at your meaning - it seems that you're suggesting cutting the line voltage to 60 vac?
 
#5
Hi,

Yes. It shouldn't be difficult to design an amplifier than does 10W or
so that with halving the the B voltage still works at much lower
power, and you do that by wiring the primaries in series.

Yes you would design the amplifier to run properly at full rail
voltage, but many simple designs would still work at low
power if you halved the rail voltage, i.e. basically crudely.
Still in "low power" mode the valves would last ages.

rgds, sreten.
 
Last edited:
#6
I guess i did understand after all. I am going to ask one more question because i'm still thinking that it might be cool (not easy or cheap, but cool) to have the seperate power stages.

Follow-up question:
I have already mounted a 12ax7 and a 6bg6 - the 6bg6 has an output impedance of 4.5k
I also have a 12bh7a that has an output impedance of 5k.

It occurs to me that these impedances are pretty close and that i might be able to make one output transformer work for both. If i understand it right then one would be loaded properly and the other would be either hot or cold.
Is a 5k output transformer going to be okay for both of those tubes?

I figure - since the 12bh7 is such a low gain tube - i'll probably end up close to the submini range with that one. That would be great for me. Then if i need more headroom i could switch to the 6bg6. Also great.

Alternative idea - to get a 4.5k OT. Match the 6bg6 properly and make the 12bh7 run a little bit hot.

Thoughts?
 
#7
Alternative idea - to get a 4.5k OT
Click to expand...

You will not see a difference between the 4.5 K OPT and a 5 K OPT. Your speaker will go from over 20 ohms near resonance to below 8 ohms (assuming an 8 ohm speaker). The OPT just reflects this load to the tube which sees somewhere between 4K and 12 K with the 5 K OPT.

I have built a dual power amp in a similar manner except mine was push pull. I used a 6BX7 for the low power mode and a pair of 6L6GC's for high power with a 6600 ohm OPT. The 6BX7 is a steroid injected 6SN7 for TV vertical sweep duty. A good 6SN7 will work too.

I simply wired both output tubes in parallel using the same driver (separate coupling caps) and OPT. The amp used fixed bias. Switching was accomplished by applying a high negative bias to the tube to be shut off. This works well in my case, but the unused tube will still see the entire plate voltage swing from the other tube. A pair of 6L6GC's driven to saturation can see over 1KV on the plates with a 350 volt B+ voltage. The 6BX7 can eat this with no problem, but some older 6SN7's will spark out.

The 12BH7A is also rated for TV vertical sweep and is a good choice for this application. It can eat +1500 volts on its plate and -250 on the grid. You will need about -50 volts to shut it completely off. The 6BG6 is a 6L6GA modified for TV sweep use. It can eat +6600 volts on its plate and -300 volts on its grid. You will need about -70 volts to shut it completely off, maybe a bit more if triode wired.

I used 4 bias pots, one for each tube (or half tube in the case of the 6BX7). Wire them up in the usual manner except that the resistor that usually goes from one end of the pot to ground goes to the center terminal of a switch. One end of the switch goes to ground, the other end goes to the negative supply which was about -100 volts in my amp (from the purple wire on a Hammond power transformer).

My amp was about 2 watts through the 6BX7 and 30 through the 6L6GC's. SE would get you about 1/2 and 10 with similar tubes.

I have another amp that used cheap TV tubes for about 1 watt in low power, and 8 watts in high power. The change was done by switching from SE to P-P.
 
#8
tubelab.com said:
You will not see a difference between the 4.5 K OPT and a 5 K OPT. Your speaker will go from over 20 ohms near resonance to below 8 ohms (assuming an 8 ohm speaker). The OPT just reflects this load to the tube which sees somewhere between 4K and 12 K with the 5 K OPT.

I have built a dual power amp in a similar manner except mine was push pull. I used a 6BX7 for the low power mode and a pair of 6L6GC's for high power with a 6600 ohm OPT. The 6BX7 is a steroid injected 6SN7 for TV vertical sweep duty. A good 6SN7 will work too.

I simply wired both output tubes in parallel using the same driver (separate coupling caps) and OPT. The amp used fixed bias. Switching was accomplished by applying a high negative bias to the tube to be shut off. This works well in my case, but the unused tube will still see the entire plate voltage swing from the other tube. A pair of 6L6GC's driven to saturation can see over 1KV on the plates with a 350 volt B+ voltage. The 6BX7 can eat this with no problem, but some older 6SN7's will spark out.

The 12BH7A is also rated for TV vertical sweep and is a good choice for this application. It can eat +1500 volts on its plate and -250 on the grid. You will need about -50 volts to shut it completely off. The 6BG6 is a 6L6GA modified for TV sweep use. It can eat +6600 volts on its plate and -300 volts on its grid. You will need about -70 volts to shut it completely off, maybe a bit more if triode wired.

I used 4 bias pots, one for each tube (or half tube in the case of the 6BX7). Wire them up in the usual manner except that the resistor that usually goes from one end of the pot to ground goes to the center terminal of a switch. One end of the switch goes to ground, the other end goes to the negative supply which was about -100 volts in my amp (from the purple wire on a Hammond power transformer).

My amp was about 2 watts through the 6BX7 and 30 through the 6L6GC's. SE would get you about 1/2 and 10 with similar tubes.

I have another amp that used cheap TV tubes for about 1 watt in low power, and 8 watts in high power. The change was done by switching from SE to P-P.
Click to expand...

So basically i'm hearing it's been done and it works well? 🙂

I like this idea. I'm still shaky about how the biasing is supposed to work but i'm learning a little more every day. I'm pretty sure i understand. If you put enough negative voltage on the grid it prevents conductance from cathode to anode, right? (Or should i say anode to cathode?)

Thank you - thanks everyone i should say.
 
#9
If you put enough negative voltage on the grid it prevents conductance
Click to expand...

That's all you need to know. Enough negative voltage shuts the tube completely off. Keep in mind that you still have audio signal on the grid, so you need enough negative voltage to keep the tube off even when the audio signal is trying to turn it on. I had about -100 volts applied to the off tube which was enough even when cranking the amp to 11 with a pedal board.
 
#11
Related topic:
I've just bought a cheap solid state amp that i'm planning to use for parts. Input jack, speaker, etc...

But i've had another bad idea - i thought to myself "i wonder if i can integrate the guts of this solid state amp into my tube amp build?"

I wouldn't want to make the signal go through the solid state circuitry all the time but it might be cool if i could flip a switch and make the audio signal detour through the solid state circuitry. In that way it would be more of a hybrid.

So like this:
Channel 1: guitar to tube pre-amp to big power tube(6bg6)
Channel 2: guitar to tube pre-amp to small power tube (12bh7)
Channel 3: guitar to solid-state circuit to tube pre-amp to big power tube (6bg6)

Dream big, right?
 
#14
Hi,

One output transformer may work for both, but SE it
will be expensive and you have no chance of 20W SE.

Realistic pragmatism is better than unrealistic dreams.

Your trying to run before you can walk. There are
shedloads of classic valve guitar amplifier schematics
available online and some of them are really good.

YMMV but my Peavey mk111 bass head is really
good, but from my hifi perspective i don't get
anything about the way it is designed.

Sometimes its best not to know with a good amp.

rgds, sreten.
 
#15
sreten said:
Hi,

One output transformer may work for both, but SE it
will be expensive and you have no chance of 20W SE.

Realistic pragmatism is better than unrealistic dreams.

Your trying to run before you can walk. There are
shedloads of classic valve guitar amplifier schematics
available online and some of them are really good.

YMMV but my Peavey mk111 bass head is really
good, but from my hifi perspective i don't get
anything about the way it is designed.

Sometimes its best not to know with a good amp.

rgds, sreten.
Click to expand...


I am certain that you're right. I've been simplifying my plans as I sort through my parts. I can see that your initial suggestion is the simplest but I'm not sure how the output would be. I've found a significant number of additional parts during the course of the week and I'm trying to figure out my best method to proceed.

I have been studying biasing a lot and I think that I have some understanding of how that should be done. I like single-ended, cathode biased amplifiers a lot so that is what I will build. From researching VVR circuits I understand that changing the B+ voltage will affect the bias of all components proportionally. I think that I like that idea best. I don't feel that I need a fully adjustable VVR circuit. It would be sufficient for me to just have a low watt and a high watt setting.

Is it safe to say that I could bias a simple Fender Champ style circuit, but instead of a 2 position power switch I could have a 3 position switch, with the third position adding a large resistor in series with the B+ voltage divider that goes to the plates?
 
Status
Not open for further replies.
Top Bottom