This is a conversation that started under a separate topic but I didn't want to slide the OP topic. I have rewritten the original questions for clarity sake.
I don't understand how using a 16 Ohm tap with an 8 Ohm speaker improves the sound. I'm not criticizing, just questing how this works to advantage. (It was mentioned that this causes tube distortion)
Another question: If you were to connect the 16 Ohm tap to an 8 Ohm speaker would that damage the output transformer?
silverfox.
Stevem
...Messing with different loads on different taps can make for too much current being applyed and making for shorted out windings, but in the way you stated of running the OT the higher load does not let the output tubes make there intended current output for the power supply you have in use.
A sure way to kill any OT is to have it working into a open load, or too high of a resistance.
In this case a OT acts like a automotive spark coil, and since the audio voltage swing rides on top of the B+ voltage there is a big voltage arc that has no place to but to punch thru the winding insulation to find a way to ground itself out.
It's a very good safty tip to place a 250 ohm 10 watt resistor across the OTs highest impedance tap a ground just in case of this open circuit issue!
Steve I get that part now, thanks.
silverfox:
Another question:
When changing the tap on the output transformer, is it correct to think of the impedance across the output transformer as analogous to a see-saw? As one side changes in a particular direction the other side changes inversely?
silverfox.
Output transformer impedance question
Moderators: pompeiisneaks, Colossal
Re: Output transformer impedance question
Probably not, except in rare cases. A tube output circuit can usually tolerate a 100% mismatch. Max power will not be transferred to the speaker (load). And sound will probably be a bit different.Another question: If you were to connect the 16 Ohm tap to an 8 Ohm speaker would that damage the output transformer?
No. The relationship is directly proportional. A lower impedance on the secondary will reflect a lower impedance to the primary. A Higher impedance on the secondary will reflect a higher impedance to the primary.When changing the tap on the output transformer, is it correct to think of the impedance across the output transformer as analogous to a see-saw? As one side changes in a particular direction the other side changes inversely?
Re: Output transformer impedance question
A transformer's impedance is based on the turns ratio. The transformer itself has no idea if it's 8K, 4K, 2K, whatever. All it knows is that a turns ratio of 25:1 is the functional equivalent of 5000:8 or 2500:4 or 10000:16. Depending on other winding parameters, it may or may not be fully interchangeable. I'm going to guess most guitar amp OT's will be OK with this treatment.
Re: Output transformer impedance question
It's not quite true that mismatching to a lower impedance improves the sound, as that depends on the valves, on the amp and on the listener.
For particular valves in a particular amp there is an optimal load that will give the maximum output power, and this corresponds to the load line graphically going through the knee of the Vg=0 curve on the plate characteristic curves. A basic generalisation is that going higher impedance than that will give more screen compression before grid clipping and change the balance of harmonics one way. Going lower impedance moves the load line above the knee and gives less screen compression (very quickly becomes none) and changes the balance of harmonics the other way. Which one sounds better actually depends on what the listener likes more for that particular amp. Both are valid directions.
Even how far you can mismatch before something is damaged depends on the amp and the bits used. Too high an impedance draws more screen current and without the right screen resistors (or none in some amps) and you burn out the screens. Too low an impedance moves the load line slope closer and more over the max plate dissipation. Push that too far without compensating with a cooler bias and you get redplating and burn out the plates. This is often well before the OT is at risk from overvoltage, but of course depends on how marginal the OT design is from operation in service.
Your first two questions are complex and don't have yes/no answers. The third question about visualising it as a see-saw is definitely one way to think of what the OT is doing.
For particular valves in a particular amp there is an optimal load that will give the maximum output power, and this corresponds to the load line graphically going through the knee of the Vg=0 curve on the plate characteristic curves. A basic generalisation is that going higher impedance than that will give more screen compression before grid clipping and change the balance of harmonics one way. Going lower impedance moves the load line above the knee and gives less screen compression (very quickly becomes none) and changes the balance of harmonics the other way. Which one sounds better actually depends on what the listener likes more for that particular amp. Both are valid directions.
Even how far you can mismatch before something is damaged depends on the amp and the bits used. Too high an impedance draws more screen current and without the right screen resistors (or none in some amps) and you burn out the screens. Too low an impedance moves the load line slope closer and more over the max plate dissipation. Push that too far without compensating with a cooler bias and you get redplating and burn out the plates. This is often well before the OT is at risk from overvoltage, but of course depends on how marginal the OT design is from operation in service.
Your first two questions are complex and don't have yes/no answers. The third question about visualising it as a see-saw is definitely one way to think of what the OT is doing.
-
stretch2011
- Posts: 501
- Joined: Tue Oct 21, 2014 2:53 am
- Location: ohio
- Contact:
Re: Output transformer impedance question
Just wanting to throw something out there and use 5881's as an example. A pair can run At 6.6k or 3.8k primaries. If you have a 3.8kish primary and instead of using the suggested 8ohm load wouldn't a 16ohm speaker cause the primary to act as a 7.2ish
Or would it be a 4 ohm load? My math may be backwards.
This is just going off what phil said.
Or would it be a 4 ohm load? My math may be backwards.
This is just going off what phil said.
A bit of a slide
Thanks katopan and others. This is enlightening.
In regards to screens and compression- My understanding is the screen catches stray electrons for the purpose of keeping the cloud at a minimum. It reduces noise.
Sounds like screen compression is, or is related to current flow through the screens. Less flow means more congestion and perhaps a higher voltage on the screen resistors- Higher heat too.
edit- Screen resistors are going to change the capacitance of the tube and alter the frequency dynamics now that I think of it. I think...
Is this how the screen resistors affect the tone. I'm probably restating what has already been said but only for the purpose of clarifying my understanding.
silverfox.
In regards to screens and compression- My understanding is the screen catches stray electrons for the purpose of keeping the cloud at a minimum. It reduces noise.
Sounds like screen compression is, or is related to current flow through the screens. Less flow means more congestion and perhaps a higher voltage on the screen resistors- Higher heat too.
edit- Screen resistors are going to change the capacitance of the tube and alter the frequency dynamics now that I think of it. I think...
Is this how the screen resistors affect the tone. I'm probably restating what has already been said but only for the purpose of clarifying my understanding.
silverfox.
Re: Output transformer impedance question
Okay - Well you shouldn't really hijack someones else's thread until they've got the answer they were seeking.stretch2011 wrote:Just wanting to throw something out there and use 5881's as an example. A pair can run At 6.6k or 3.8k primaries. If you have a 3.8kish primary and instead of using the suggested 8ohm load wouldn't a 16ohm speaker cause the primary to act as a 7.2ish
Or would it be a 4 ohm load? My math may be backwards.
This is just going off what phil said.
So anyway a bigger impedance on the secondary winding will mean a bigger reflected load to the primary.
He who dies with the most tubes... wins
-
stretch2011
- Posts: 501
- Joined: Tue Oct 21, 2014 2:53 am
- Location: ohio
- Contact:
Re: Output transformer impedance question
Damn! I thought I was helping lol
I don't know about sounding better because that is opinionated but for in the example I gave you would be louder.
I don't know about sounding better because that is opinionated but for in the example I gave you would be louder.
In this case More is Better
Perhaps due to the nature of the original posting and that I've slid the topic a little by bringing up the screen resistor affect on the tone.
Stretch2011- "If you have a 3.8kish primary and instead of using the suggested 8ohm load wouldn't a 16ohm speaker cause the primary to act as a 7.2ish"
That's what I would think.
I'm beginning to have enough questions on the output section now it may deserve it's own thread. I'll do some searching first. I don't know what I don't know so conversations like this open doors that lead to more questions. The "Z" Files? Da da da da da da....
silverfox.
Stretch2011- "If you have a 3.8kish primary and instead of using the suggested 8ohm load wouldn't a 16ohm speaker cause the primary to act as a 7.2ish"
That's what I would think.
I'm beginning to have enough questions on the output section now it may deserve it's own thread. I'll do some searching first. I don't know what I don't know so conversations like this open doors that lead to more questions. The "Z" Files? Da da da da da da....
silverfox.