Thanks, I understand now.
I thought DF96 meant like in Class B, when the tube is completely cut off when it reaches B+ voltage. 😱
Question,
Everyone who had an experience with these chinese amps says that they are biased too hot and will result in short tube life.
When biased hot, does that mean the idle grid voltage is more negative or less negative?
Question Example:
Hot -12v, Cold -24v.
Correct?
I thought DF96 meant like in Class B, when the tube is completely cut off when it reaches B+ voltage. 😱
Question,
Everyone who had an experience with these chinese amps says that they are biased too hot and will result in short tube life.
When biased hot, does that mean the idle grid voltage is more negative or less negative?
Question Example:
Hot -12v, Cold -24v.
Correct?
Last edited:
This means that the bottom half of the waveform will be compressed (like the grid graph) when the grid reaches 0V, while the top half still complete and intact, resulting in an asymmetrical compression?
Not the best for Hi-Fi is it?
What would be the balance point between hot and cold for the most symmetrical reproduction of the waveform?
Not the best for Hi-Fi is it?
What would be the balance point between hot and cold for the most symmetrical reproduction of the waveform?
Last edited:
I found the Triode/Pentode Loadline Calculator on other thread which is very helpful.
Well I feel I've come a long way past these several days gathering and receiving info.
When the amp arrives I'll do comprehensive voltage measurements, including transformer turn-ratio to find the primary impedance.
Oh, and most importantly, listening tests with different output modes once the mods are done.
Well I feel I've come a long way past these several days gathering and receiving info.
When the amp arrives I'll do comprehensive voltage measurements, including transformer turn-ratio to find the primary impedance.
Oh, and most importantly, listening tests with different output modes once the mods are done.
Last edited:
Yes, but this lop-sided swing is what produce the even-order harmonic profile which for some is pleasing to listen to.
You've answered this yourself. You need to find the bias point where the loadline crosses the least compressed grid lines. However, in the case of single ended amp, you can't play much with the bias point or you'll sacrifice output power (too cold) or tube reliability (too hot). Hence, it comes back to a matter of tube choice: find the tube with the most linear characteristic. For some, this means triode-wired pentode. For example, look at the beautifully linear triode-wired 4P1L characteristic here: 4P1L triode SPICE model | Bartola Valves
Thanks for posting the schematic. You are correct, there appears to be no choke in this little amp.
The doubled up input triode is a pity.. but by doing that the designer can get away with a much smaller load, which is probably necessary due to the power supply.
I don't know why chinese designers don't build better power supplies so they can then build better amps (Ok I do know why - it all comes down to the allmighty $).. They could do a voltage doubler.. then use choke input.. use the 2nd triode in the envelope for something useful... but of course then they might need a few more parts. It might also start to look like a nice old Marantz or Fischer - perhaps we should not go there today though. 😉
EL34 is honestly OK in pentode. I don't like it triode strapped though. Not every pentode sounds better when triode strapped.
The doubled up input triode is a pity.. but by doing that the designer can get away with a much smaller load, which is probably necessary due to the power supply.
I don't know why chinese designers don't build better power supplies so they can then build better amps (Ok I do know why - it all comes down to the allmighty $).. They could do a voltage doubler.. then use choke input.. use the 2nd triode in the envelope for something useful... but of course then they might need a few more parts. It might also start to look like a nice old Marantz or Fischer - perhaps we should not go there today though. 😉
EL34 is honestly OK in pentode. I don't like it triode strapped though. Not every pentode sounds better when triode strapped.
Consult the valve datasheet, or draw load lines.James Freeman said:
My point about 300V peak from 300V rails was that you can't get a 300V peak AC signal from 300V DC rails, but that is what you seemed to be assuming in your impedance and power calculation. Depending on how compressed you are happy for the peak to be, you might get 270V peak or 250V peak.
Hi,
The simple facts are a good valve hifi amplifier doesn't sound
like a solid state, especially when they are both clipping.
Trying to add a plethora of options based on experience with
guitar amplifiers is misguided. The amplifier should do the
hifi job best in the stock state in arrives in, wired up to
conventional hifi speakers, which are not improved
in any way by any form of high impedance drive.
rgds, sreten.
http://www.lundahl.se/easy-to-build-se-amplifier/
The simple facts are a good valve hifi amplifier doesn't sound
like a solid state, especially when they are both clipping.
Trying to add a plethora of options based on experience with
guitar amplifiers is misguided. The amplifier should do the
hifi job best in the stock state in arrives in, wired up to
conventional hifi speakers, which are not improved
in any way by any form of high impedance drive.
rgds, sreten.
http://www.lundahl.se/easy-to-build-se-amplifier/
Last edited:
Hi,
Especially because a SE tube amp is never linear at any significant power, it is always deforming the waveform.
According to the Triode / Pentode Loadline Simulator, there is no region which is distortion free.
The linear regions is so small in SE configuration (one tube) that no significant power can be achieved without waveform deformation.
Last edited:
Depends what you consider to be distortion free or not, and what significant power is defined as.
I suggest you learn how to draw a load line the old fashioned way, and post what you think it is here. No worries - even though most of us sound a bit old and crotchety in our replies, we do really mean to help! 🙂
If you look carefully at the photo of the amp you have coming, its output tube is actually Shuguang EL34-B. The EL34-B is not a pentode but a beam power tetrode (more similar to a 6CA7 than an old fashioned EL34). Its power ratings as stated are slightly higher, so maybe that helps. But any comments related to EL34 doesn't really apply.
If you're planning to mess with the circuit, which can be fun for its own purposes (if you know how to avoid electrocuting yourself, of course) I strongly recommend putting together a SPICE model of it using the (free) LTSPICE application and some 6CA7 and 6SL7 device models you can download from the web. A much more efficient and safer way to try things out before more time consuming and possibly destructive physical circuit changes. Very educational, too. I think I have a SPICE file that would be pretty close if you're interested in going the geeky route.
The bit about not operating the amp without a load is real, even with no input signal. When (and I don't mean "if") something goes against expectation, a large AC peak can occur between windings and that will be the end of it for the transformer if there's nothing to limit it. A 47 ohm resistor on the 8 ohm secondary won't cost much power and would save you from having an expensive doorstop instead of a project amp.
Just some friendly advice from someone whose been there a little before.
If you're planning to mess with the circuit, which can be fun for its own purposes (if you know how to avoid electrocuting yourself, of course) I strongly recommend putting together a SPICE model of it using the (free) LTSPICE application and some 6CA7 and 6SL7 device models you can download from the web. A much more efficient and safer way to try things out before more time consuming and possibly destructive physical circuit changes. Very educational, too. I think I have a SPICE file that would be pretty close if you're interested in going the geeky route.
The bit about not operating the amp without a load is real, even with no input signal. When (and I don't mean "if") something goes against expectation, a large AC peak can occur between windings and that will be the end of it for the transformer if there's nothing to limit it. A 47 ohm resistor on the 8 ohm secondary won't cost much power and would save you from having an expensive doorstop instead of a project amp.
Just some friendly advice from someone whose been there a little before.
Cool! I was beginning to think I've come to the wrong place. 🙂
First, I'm going to measure all voltage points of my real amp (yet to arrive), and I'm going to find out the turn ratio of the transformers (1vac at the secondary, measure at the primary without tube).
After that I will be able to draw a proper load line with real B+ voltage and primary impedance.
No worries, I've got my DMM and 100K 5W two sided crocodile cable ready to discharge those caps.
First mod is to install a bleeder resistor across the caps if there isn't one.
The statement about the high distortion with SE amps is straight form what I see with the Tube Loadline Simulator.
They shipped it with "PSVANE HIFI serie EL34" which is probably a Shuguang.
No problem, I've been using LTSPICE for many years and simulated SS and Tube amps.
Any tips on the transformer simulation (resistance etc...) or should I stick to ideal?
No, I mean 270V peak AC voltage. This could be 30V or 570V at the anode, from 300V supply rail.James Freeman said:
That is why people who want significant power and hi-fi generally use push-pull.
Okay now I understand.
What do you think will improve the DC stability and provide more current when needed? Where to remaining 30V go?
The Tube rectifier definitely has no place in a hi-fi amplifier.
Bigger value caps?
I also guess that the Choke can slow the current when the amp needs it the most and sag the voltage, like a soft start... I might be wrong.
EDIT:
jazbo8 said:
tonequester said:
mike567 said:
I believe mike567.
Okay, the Choke is not the cause for voltage sag in class A SE, and we get all the benefits.
Me: Great discussion James.
Me: Thanks!
Last edited:
For class A operation, the current is nearly constant, so there's not going to be much sag, but the choke can certainly help to get the ripple voltage down further, if you have the space for it.
Thanks jazbo8,
I wonder why google provides all this info mostly from diyaudio... 😉
P.S
My thread has been rated "Terrible" but I love it anyway. 🙂
Why the 30 min edit disability ???
I wonder why google provides all this info mostly from diyaudio... 😉
P.S
My thread has been rated "Terrible" but I love it anyway. 🙂
Why the 30 min edit disability ???
Last edited:
There's a spreadsheet somewhere on the web that can generate a SPICE model for an output transformer given its impedance ratio (turn ratio squared), winding resistances and bandwidth. I got reasonably accurate results from it. If googling doesn't turn it up for you, I'll see if I can find it on my puter.
Well you may have started the thread, but you can not control its content, so the rating isn't a reflection on you personally anyway...🙂
Or do what jackinnj suggested here - use ideal transformers and add all the parasitic components separately.
Fellows, any procedure of warming up the tubes?
Should I wait a few seconds while the filaments heat up?
Although it should not matter because in SE amp the current through the tube is max at idle, and the HT and filaments turn on together.
* Rectifier tube will be gone.
According to valve wizard, cathode stripping should not occur in "low voltage" tubes we use for amplification.
Should I wait a few seconds while the filaments heat up?
Although it should not matter because in SE amp the current through the tube is max at idle, and the HT and filaments turn on together.
* Rectifier tube will be gone.
According to valve wizard, cathode stripping should not occur in "low voltage" tubes we use for amplification.
Last edited:
- Status
- Not open for further replies.