I think I've found my answer. The grid choke's impedance forms a voltage divider with the output impedance of the preamp tube, not the grid stopper. In this case, 7700 for the 6SN7GT and the 44k of the grid choke. As the frequency rises so does the impedance so that there is always a loss of gain largest at the lowest frequencies and decreasing as frequencies rise. At 20hz there is approximately a 15% loss in gain.
The drawback of the grid choke is that you can't overcome this problem with more gain as the relative gain is always going to suffer in the low frequencies. I now understand that there three options for using a grid choke. First, to use a very low output impedance preamp tube. Or second, an enormously large inductance choke. Or a combination of the two.
The drawback of the grid choke is that you can't overcome this problem with more gain as the relative gain is always going to suffer in the low frequencies. I now understand that there three options for using a grid choke. First, to use a very low output impedance preamp tube. Or second, an enormously large inductance choke. Or a combination of the two.
grid choke dc resistance matters too, it forms part of the equation...
consider the tube as a thevenin voltage source, and you will see how it goes...
true so you design around this....
consider the tube as a thevenin voltage source, and you will see how it goes...
true so you design around this....
Are you considering that the inductance is frequency dependent?
Features of Valab GC640-5 filter choke
* High grade thin Z11 lamination.
* 5n oxygen free copper wires was used in coil wiring.
* 20000 turn coils, generating 640H at 120Hz and over 2500H at 20Hz.
This is according to the manufacturer. These are on Ebay at 2 for $50. Just search for grid choke. I have used these and see zero drop off in the bass. FYI
The manufacturer states 2500H @ 20 Hz and 640H @ 120 Hz or
314K @ 20Hz and 482K @ 120 Hz
Features of Valab GC640-5 filter choke
* High grade thin Z11 lamination.
* 5n oxygen free copper wires was used in coil wiring.
* 20000 turn coils, generating 640H at 120Hz and over 2500H at 20Hz.
This is according to the manufacturer. These are on Ebay at 2 for $50. Just search for grid choke. I have used these and see zero drop off in the bass. FYI
The manufacturer states 2500H @ 20 Hz and 640H @ 120 Hz or
314K @ 20Hz and 482K @ 120 Hz
grid chokes are fine to use on tubes where bias runaways are a possibility...
with its low dc resistance compared to a mere resistor...
with its low dc resistance compared to a mere resistor...
something does not add up, inductance is not frequency dependent, inductive reactance is....
z11 is the equivalent of M6, and is 0.35mm thick laminations....is the DC resistance quoted as well?
Inductance is a function of the mu of the core - higher flux normally leads to higher mu. Up to saturation that is. Lower frequencies give rise to higher flux levels in the core, given same amplitude drive.
Right. The math doesn't make any sense. Plus they are cheap and from Taiwan.
DC resistance on mine is 2.68k a fraction of the 220k in the schematic. The chokes I have are capable of much greater current than necessary I doubt they will ever come close to saturation.
DC resistance on mine is 2.68k a fraction of the 220k in the schematic. The chokes I have are capable of much greater current than necessary I doubt they will ever come close to saturation.
Last edited:
Those measurements of increasing inductance vs decreasing freq look totally reasonable to me. Where's the dodgy math?
So do you add the DC resistance to the AC impedance? That would make the total resistance 46,680 and would have a slightly more positive effect on the voltage divider?
No - AC impedance is at right angles to resistance. So you need a vector addition, not a simple scalar addition.
I do not agree. You cannot ignore DC resistance, it´s always there.
If it wasn´t, nobody would waste time with (for example capacitor esr effects).
AJT: The DC resistance on those ebay chokes is 20.2K. I just measured one.
Further, I don't see anything wrong with the quality of them. I have now used six of them with no problems and no stray humm pickup. For $25 each I would say they are an incredible value.
Further, I don't see anything wrong with the quality of them. I have now used six of them with no problems and no stray humm pickup. For $25 each I would say they are an incredible value.
From my understanding a simple resistor effects both AC and DC. The impedance is based on the inductance of the choke. But there is also the resistance of the wire itself that's why I thought they should be added together?
I spoke to Arlen at Heyboer this morning. He is not aware of any inductance change with frequency low or high. Doesn't mean it isn't there though.
As for the Thevenin calculation I'm working on that.
I spoke to Arlen at Heyboer this morning. He is not aware of any inductance change with frequency low or high. Doesn't mean it isn't there though.
As for the Thevenin calculation I'm working on that.
If you look at the SAC Thailand grid choke page they claim 7000H @ 12 Hz and state 5M resistance @ 1000 Hz which is about 800H so according to them their grid chokes are:
7000H @ 12 Hz
800H @ 1000 Hz.
SAC Thailand
7000H @ 12 Hz
800H @ 1000 Hz.
SAC Thailand
i have built a dozen of set amps with grid choke in place of grid resistor and i always prefer the more open and dynamic sounds with the choke .
watch out with the Silk choke the wire is very thin and if you have hands to male donuts likes mine you could easily break them with too much heat on them..
the only case i don't put grid choke is when i must put a blocking capacitor before . CL filters sometimes are more unpredictable than CR ones 🙂
watch out with the Silk choke the wire is very thin and if you have hands to male donuts likes mine you could easily break them with too much heat on them..
the only case i don't put grid choke is when i must put a blocking capacitor before . CL filters sometimes are more unpredictable than CR ones 🙂
He didn't suggest ignoring resistance; he suggested adding it in correctly (in quadrature) instead of wrongly (in phase).hpeter said:
Well If my math is correct, (always suspect) 74/7000=.0106ma. My chokes are 39 gauge wire which is good for about 17ma. So the SAC chokes are probably a lot less. Especially considering their size and induction spec.
Vive la France!
Vive la France!
Do they specify the choke self resonance?
I greatly suspect it may lie well within the audio band, given the large inductance *and* distributed capacitance.
20000 turns of wire must have large parasitic capacitance.
Wah Wah effect anybody?
Not kidding, please somebody measure and post that important parameter; I have none available or would measure that myself.
I greatly suspect it may lie well within the audio band, given the large inductance *and* distributed capacitance.
20000 turns of wire must have large parasitic capacitance.
Wah Wah effect anybody?
Not kidding, please somebody measure and post that important parameter; I have none available or would measure that myself.
As far as I can see I haven't said to ignore DC resistance. Performing a vector addition as I suggest would specifically meaning including DC resistance in the calculation, the polar opposite of ignoring it.
@JMFahey - very good point. Most probably the self-resonance is going to be well inside the audio band so cannot be ignored in impedance calculations. My guess based on winding trafos with a few thousand turns is it'll look capacitive before 4kHz, perhaps well before.
Last edited:
way i see it, grid chokes are not meant to pass dc currents, if you can post a scheme on what you actually intend to do, then we can understand better....
looking at your post #8 i can see that you are on the right track, chokes do indeed provide a lower dc resistance to ground than any resistor would without loading down the drivers hard....
if your grid will not draw current with signal peaks or close to 0v grid voltage, then the choke is okey,
you other concern is how the choke loads the driver stage as gain and output of that stage is impacted
by the choke impedances, now with chokes what we are looking at is the choke impedance at the
lowest frequency of interest like say 30hz, at much higher frequencies, impedance of the choke will be way way high...
so higher frequencies are not much of a concern until choke parasitics come into lay...
- Status
- Not open for further replies.