Bias measuring resistor(s)

[txstrat]

Hi all,

while studying the Fender Hot Rod Deluxe schematic I stumbled over the 1 ohms resistor at the power tubes cathodes paralleled by a 1N5062 diode.
I ususally use two separate 1 ohms 5% 2w resistors (unfortunately I haven't found a source for 1% resistors) and no diode.
I take the readings from both the cathodes (usually differing by 5-8 mv) and set the bias for the average value.
Now I think the 1 resistor design would spare me the average value thing. I could take one reading and set the bias.
What do you guys think about the two or one resistor design?
What is the diode in the HRD for?

thanks

Matt

[Enzo]

2 watt 1 ohm 1% resistors?

Mouser 71-CPF21R0000FKE14 62 cents
Mouser 71-RS02B1R000FE70 66 cents

Digikey WHC1R0FET-ND 88 cents.

SOme were Vishay, some were Ohmite, and if you can;t find 1% 2w, look at 3w, or 2%. Bias isn;t that critical anyway.

I know US suppliers shiiping to Germanywouldn;t be efficient, but if most of our large parts houses have them, isn't there a large parts house or two over there? RS maybe?

I see no problem with a single resistor, especially if you don;t have individually adjustable bias anyway. it all depends how much noodling you plan to do. If you want to know what each tube is doing, then spend the extra buck and mount individual resistors. If you are concerned that one tube may be way off from the other even though the total seems reasonable (like one really hot tube paired with one really cool tube) then mount two resistors. If you are not installing odd tubes, then use one. I generally install matched sets of tubes in amps. Very rare would I repair someones amp by sticking one new tube in with old ones in the power stage.


And the diode? SInce it is forward biased, and parallel to that 1 ohm, it wonl;t be doing anything until the voltage across the 1 ohm resistor gets over the juction voltage of the diode, roughly half a volt. Well half a volt is 500ma through the tubes, so they are in serious trouble if the diode has to come on. SO if I had to think about it, I'd say the diode is there to protect the resistor in case of tube failure. And note the resistor is a flame proof type, (Your metal film or wirewound types would be OK.) so they have that on their mind.

[Robert M. Martinelli]

Hi Matt,
I always prefer to have one of them for each tube, this way I can always detect severely mismatched tubes.

With only one bias measuring resistor in, all you know about the "X mV reading" is that you have "X mAmps flowing to ground", but you have no clue about each tube's contribution. If you're lucky and you happen to get a reasonably matched pair or quartet ( as you said, 5-8 mAmps ) everything is OK, but what if your tubes are "miles away" in terms of quiescent current ?

When there's the chance to make things better at such a low cost why hesitate? I understand that at FMIC cost saving is an issue when producing many runs of, say, 10000 units each ( and by saying "understanding" I don't mean "approving" ) but I always choose to follow a "safe and sound" design philosophy, all the more so when hand-building things one by one.

As to the 1N5062 diode, I think that most of the time it just sits there.

If the diode is lucky and the owner lives close to a Cali beach, that 1N5062 will most likely spend its life enjoying the hot climate, doing some "sightseeing", singing "I wish they all could be California girls" and maybe drinking a beer or two .

OK, seriously.....the 1N5062 is an "avalanche diode", and you can normally find it ( reverse biased ) used as a surge protection or a voltage clamper ( similarly to a Zener diode ). Anyway, this doesn't seem the case, as the diode is forward biased.

In normal conditions the voltage at the 1N5062 anode is way lower than its forward voltage, so "circuit-wise" it's just like the diode is not there, as it would take over 600 mAmps ( 300 mAmps from each tube ) for the voltage at the 1N5062 anode to reach the forward voltage ( 0,6 V at 25°C ) and turn the diode into the ON state.

All in all, ( and I could be wrong ) I think that CR3 is only there to provide an alternate ( and fast ) path to ground should the 1 Ohm resistor fail.

JM2CW

Cheers

Bob

Edit - Oooops, Hi Enzo.... another "simulpost" ( I think you should start to ask royalties.... )

Anyway...I just saw your post and have to say your explanation is better than mine ( it makes more sense )

[txstrat]

Thanks for your fast replies guys.
Well, I always buy tubes in matched pairs therefore they should not be way off each other. That would spare me one resistor and that noodling around with the two values. On the other hand I would know IF they are way off. Anyway, great to have some other opinions.
I think I have to have a closer look for another online supplier for 1% or 2% resistors.

Ahem, 20 seconds later: I've found the right ones at my supplier. Just a mouseclick on the right link.

[Enzo]

I am all in favor of explaining the same thing more than one way - it is more likely to sink in at the other end.

I find your take on the purpose interesting. I think it protects the resistor, you think it covers the resistor's function. Hard to say. I think if the 1 ohm resistor fails, having the diode there to keep the tubes running is the least of our worries. as we both exampled, it takes a ton of current to turn onthe diode, and that much would probably take out the resistor.

I think Fender used just one because they wanted a simple bias adjustment point, rather than a diagnostic tool. "Set bias for 60mv at TP1." With two test points, it would turn into, "Measure current at TP1 and TP2, take average reading of the two, and adjust bias until they average out to 60mv."

As much as people obsess over bias, some factories set the amps by mains draw. The spec reads, "set bias so mains current draw at idle is 72 watts" Or whatever. Like an old Fender Twin, which lacked a test point, but the bias control was accessible from outside through that chassis hole screwdriver slot. QC guy plugs the amp in, makes sure it works, watches his mains current-metered outlet, and sets the bias to the spec that way. Didn;t even have to open up the chassis.

And why not? If you plan to set your pair of tubes to 21 watts each - 70% of 30 - well that is 42 watts. Doesn;t matter if the B+ is higher or lower on a given chassis, 42 watts is 42 watts. And given that they would already know what the rest of the amp drew, that plus the 42 was the spec to shoot for.

[mooreamps]

if both resistors are "matched", you may not care if they are 1%. If you can measure the resistor, you can then calculate plate current. I don't have a print for the HotRod Deluxe, so I can not comment on the diode. I also understand it tends to be convention to measure off the cathode circuit, where as a more accurate measurement is off the plate circuit, which is the method I use.

-g

[mooreamps]

Hi all,

while studying the Fender Hot Rod Deluxe schematic I stumbled over the 1 ohms resistor at the power tubes cathodes paralleled by a 1N5062 diode.
I ususally use two separate 1 ohms 5% 2w resistors (unfortunately I haven't found a source for 1% resistors) and no diode.
I take the readings from both the cathodes (usually differing by 5-8 mv) and set the bias for the average value.
Now I think the 1 resistor design would spare me the average value thing. I could take one reading and set the bias.
What do you guys think about the two or one resistor design?
What is the diode in the HRD for?

thanks

Matt

I just pulled the schematic and the data sheet for the cathode diode.
The diode is a 1n5062 Standard Avalanche Sinterglass Diode. In this circuit, with a forward voltage up to 60 millivolts, the diode is not conducting. Thus, as used as a test point ; TP 30 in this case ; the nominal voltage recorded in the print is 60 millivolts, which would relate to a static plate current of 30 milliamps per power tube. All the plate current form both tubes is flowing through the resistor. If in the event of a tube short, the diode begins to conduct the majority of the plate current, shunting it away from the resistor.

It looks like an extra layer of insurance, since the 1 ohm cathode resistor is rated at 1 watt, and thus itself would be able to withstand up to 1 ampere of current.

The cathode resistor in this circuit, it's function is not for tube matching. I suspect it's function as a test point is used in the manufacturing test to ensure the power stage is working at the factory.


-g

[tubeswell]

Hi txstrat

1/2W ones will do fine. PM me with your snail mail details and I'll send you a few 0.6W MF ones for keepsakes.

[Robert M. Martinelli]

I just pulled the schematic and the data sheet for the cathode diode.
The diode is a 1n5062 Standard Avalanche Sinterglass Diode. In this circuit, with a forward voltage up to 60 millivolts, the diode is not conducting. Thus, as used as a test point ; TP 30 in this case ; the nominal voltage recorded in the print is 60 millivolts, which would relate to a static plate current of 30 milliamps per power tube. All the plate current form both tubes is flowing through the resistor. If in the event of a tube short, the diode begins to conduct the majority of the plate current, shunting it away from the resistor.

-g

Hi Gary,
The diode, even if an Avalanche one, here is not used in its "proper" configuration ( reverse biased ) but forward biased.

In this configuration the diode doesn't exploit its "Avalanche" behavior, and cannot act as a "clamper", but it rather behaves like a "normal" one; this means it takes the full forward voltage ( 0,6 V ) for it to start conducting, so that only currents in excess of 600 mAmps will start to flow through the diode, while anything in the 60 ( quiescent ) to 600 mAmps ( blown tube/tubes ) range will be flowing through the resistor alone IMHO. This seems reasonable enough because, as you rightfully noted, the resistor should be able to withstand currents up to 1 Amp, so a 600 mAmps threshold seems a sufficiently safe one for the diode to start conducting before the resistor suffers any damage.

JM2CW

Cheers

Bob

[mooreamps]

Hi Gary,
The diode, even if an Avalanche one, here is not used in its "proper" configuration ( reverse biased ) but forward biased.

In this configuration the diode doesn't exploit its "Avalanche" behavior, and cannot act as a "clamper", but it rather behaves like a "normal" one; this means it takes the full forward voltage ( 0,6 V ) for it to start conducting, so that only currents in excess of 600 mAmps will start to flow through the diode, while anything in the 60 ( quiescent ) to 600 mAmps ( blown tube/tubes ) range will be flowing through the resistor alone IMHO. This seems reasonable enough because, as you rightfully noted, the resistor should be able to withstand currents up to 1 Amp, so a 600 mAmps threshold seems a sufficiently safe one for the diode to start conducting before the resistor suffers any damage.

JM2CW

Cheers

Bob

Yes it is, and yes it does, and 60 milliamps across a 1 ohm resistor will have a voltage drop of : 60 millivolts, or 0.060 volts. No enough to forward bias the diode. Personally, I would have built that cathode circuit a little differently, since it does not tell you if a fault condition exists. I do have a BITE circuit for the power stage, but that lamp is always on if the tubes are off stand-by with normal plate current flowing.





-g

[txstrat]

I do have a BITE circuit for the power stage, but that lamp is always on if the tubes are off stand-by with normal plate current flowing.

You mind sharing that circuit with us? I mean do you have a schematic for it?

[mooreamps]

You mind sharing that circuit with us? I mean do you have a schematic for it?

I have it captured somewhere in one of my tube amp folders. I'll send it as soon as I finds it.



-g