DIY Audio Projects Forum

We had a buy,swap and sell night at the Melbourne audio club. The HP amp I just made based on the OPA2134 chip sold (only had it a few days, sob) but hopefully the K272B will take its place. I have been able to bend the tubes and can now fit it in my standard case upright.

_________________
Projects:"retro2308" - chip based headphone amp | ”Calibre 834” - tube phono MM preamp | ”najah” - Raw 180W Tripath Class D power amp | "Icon" - Shuguang CV-181Z preamp

http://diyaudioprojects.com/Forum/viewtopic.php?f=9&t=4234 See this link for ongoing K272B posts

_________________
Projects:"retro2308" - chip based headphone amp | ”Calibre 834” - tube phono MM preamp | ”najah” - Raw 180W Tripath Class D power amp | "Icon" - Shuguang CV-181Z preamp

Can someone help me understand the 6418 characteristics graph? I'm new to tubes, and was hoping to learn about them by starting off with this tube and circuit.

I'm confused by the nomenclature on the tube datasheet.

Page 3 (Tentative Data), top graph says "Plate Volts" vs "Plate or Grid #2- Microamperes",
whereas the bottom graph says "Plate Volts" vs "Plate Microamperes".

On all the Oatley circuits Grid2 is tied to the plate, so aren't Grid2 and the Plate volts the same? Maybe the top graph means (Plate or Grid) #2? i.e. it is just refering to Grid2? Confusing to me.

They are using subscript "c" to stand for "grid"? So that Ec1 is grid1 voltage and Ec2 is grid2 voltage?

What's Ib (I subscript b)? Current through grid1 in uA? The top graph show the first trace labeled "Ec1=0", but the rest of the solid line curves are for Ib? I.e. the secord curve labeled "-0.4" means for Ib=0.4uA?

I would use the bottom graph to draw a load line?

There's lots of info on tube characterstics I'm reading, but first I need to get this nomenclature correct...

Thanks for any help you can offer,
Keith Ostertag

Hi Keith!
I will try and answer some of the questions for you.
The data sheet that I have, which is Tungsol, seems to have some typos. I don't know what they are referencing on the charts as "b" unless they are meaning plate. In addition to "Ib", there is also "Eb", & "Ebb". Yes, they have mislabled the grids in the charts as "c", should be "g1" and "g2". In the Oatley circuits, these tubes are configured for "triode" mode, which is why plate and g2 are wired together. The references in the data sheet for g2 voltage is for when the tube is being operated in pentode mode, where g2 would be connected via a bias resistor to B+, as in the K301 phono preamp kit.

_________________
The key to a successful build is to keep the smoke IN the circuit.
-Les

We cannot solve our problems with the same thinking we used when we created them. - Albert Einstien
_________________________________
LM380 Bridged Guitar Amp, Oatley K301 Phono Pre-amp, Oatley K272 Headphone Amp, Tube proto-board

Thanks much Les! That makes more sense to me. I'm also looking at the Tungsol datasheet for the 6418- the only one I have seen.

I have also read, either here or another website, that pin3 is labeled wrong- should say "F-, G3", not "F+, G3". The correction shows you the correct polarity for the filament voltage.

So, for these Oatley circuits using the 6418 tube as a triode, I should only use the bottom graph to plot a load line, correct?

Keith

Yes. For pentode, you would use either the 2nd or 3rd chart, depending on the bias you want to use on g2.


That is correct. Pin 3 is the cathode connection, and therefore the (-) filament.

_________________
The key to a successful build is to keep the smoke IN the circuit.
-Les

We cannot solve our problems with the same thinking we used when we created them. - Albert Einstien
_________________________________
LM380 Bridged Guitar Amp, Oatley K301 Phono Pre-amp, Oatley K272 Headphone Amp, Tube proto-board

Hey Les- In trying to understand how this circuit works I am comparing it to a conventional common cathode triode circuit. But this one is different since it uses a direct heater/filament. And, there is no cathode resistor, AFAIK. What is biasing this circuit?

Also, the datasheet says "max cathode current" = .5mA. How is that different from the heater current, which they recommend to run at about 10mA?

Thanks if you can help.
Keith Ostertag

No bias resistor on cathode. I know it seems odd because g1 cannot go below k, which means 0-sig is above k. It is operating as, what I call, grounded cathode. I've seen only one other circuit operate in this mode. You can see the AF amp diagram from a circuit I traced from a Sears SilverTone AM/FM/Phono that uses a similar set-up in the driver (6SQ7) circuit:


In the 272A, you notice that V1 H/k is in series with V2 H/k, thus V2 H/k is biased above V1, and therefore the reason the VR2 is referenced to the connection between p5 of V1 and p3 of V2. I'm not sure if C3 (capacitor off p5 of V2) is there for power stability or if it is acting as bypass.


Cathode current (k(I)) refers to the current flowing between plate and cathode, where as heater current (H(I)) is just that. That means, however, you need to be aware of H(I) while designing the cathode bias to be sure you are not also putting too much voltage on the htr circuit.

The best approach for a DHT, IMHO, is to use a heater supply that is floated, that is, no ground reference at all. The method being applied in the 272 is not the best method because the heater and cathode are using a common ground. I am considering redesigning the 272 from an approach of an isolated heater supply. I'm not sure the ringing problem with these little jewels is totally mechanical, but rather possibly aggravated by biasing and loading issues. I'm sure the tubes have some inherent microphonics, but it should not be more than any other "traditional" tube. I can't see these being used in any high end product if the inherent microphonics are truly that serious, and supposedly they are used in high end foil type studio microphones and were used in military field radios. I really have hard time accepting it's purely mechanical based also on the fact my K301, which is well damped mechanically, will self oscillate from time to time even during no signal. I have seen references on other forums that support my suspicion that there is a loading issue that is contributing.

_________________
The key to a successful build is to keep the smoke IN the circuit.
-Les

We cannot solve our problems with the same thinking we used when we created them. - Albert Einstien
_________________________________
LM380 Bridged Guitar Amp, Oatley K301 Phono Pre-amp, Oatley K272 Headphone Amp, Tube proto-board

Thanks Les- you are very helpful, I appreciate it!

I'll need to think about these things... just learning.

RE: floating the heater- I had read on a different forum that since the heater is also the cathode, floating it means you'd get no plate-cathode current, i.e., one side must be returned to B- somehow. Though I'm not sure if that makes a huge difference, since the current is so small anyway- it's the potential that's amplified (gain) and transfered to the driver stage, correct? But then again. the plate current determines (in part) the operating characteristics, so...

So far, I have been looking at feeding the heater with a constant current source, and leave the cathode tied to ground as-is.

My current stumbling block is how to deal with the high output impedance of the tubes- I am hoping to use a discrete emitter follower or FET stage. Just as a personal preference, I want to use a discrete stage- not opamp.

Keith Ostertag

In general, when speaking of tubes, the heater circuit is a separate power supply from the B+. With the 6814 circuits, I'm kinda on the fence. I'd almost prefer the heaters have their own separate supply from the B+ source, while at the same time since these are low voltage tubes a single source supply has its appeal.


Simplified diagram.. Ideally, B+ and heater supplies are fully separate.

As far as gain... just keep in mind that it really depends on where in the circuit we're looking. Pre-amps are always voltage amplifiers where drivers & finals (power tubes) are current amplifiers. Some tubes do better at one or the other. For example, a 12AX7 is more suited for voltage amplification, which is why it is almost always found in a pre-amp section where as the 12AU7 is better for current and is typically found in a driver section. The 6418 is suited as a pre-amp, not a driver or final. It can amplify voltage well, but can't put current behind that.

_________________
The key to a successful build is to keep the smoke IN the circuit.
-Les

We cannot solve our problems with the same thinking we used when we created them. - Albert Einstien
_________________________________
LM380 Bridged Guitar Amp, Oatley K301 Phono Pre-amp, Oatley K272 Headphone Amp, Tube proto-board