Adjustable Fixed Bias on Geloso G1 1020 A is it possible ?

[kagliostro]

I was thinking to make adjustable (and may be to double) the bias circuit on a Geloso G1 1020 A

Thinking to the way to obtain it (at least two or three choiches) I stopped and considered that in parallel with the Bias Voltage Divider are connected the in series filaments of the preamp and PI tubes

Can I mod the Bias circuit or it will affect the energy feed to the heaters ?

I think I can Because the Resistors are of high value and we know no current flow on the Bias Path



Thanks

Franco

[sluckey]

Try this...

[HotBluePlates]

... I stopped and considered that in parallel with the Bias Voltage Divider are connected the in series filaments of the preamp and PI tubes ...

We know 12.6v heaters in series will pass 150mA, and they will split the bridge output voltage equally unless something draws enough current to upset this.

     Geloso Plan:  10kΩ + 15k5Ω = 25k5Ω --->  25.2v / 25.5kΩ = ~1mA  ---> heaters draw 150x the current, no upset

     Sluckey Plan:  25kΩ + 15kΩ = 40kΩ --->  25.2v / 40kΩ = 0.63mA  ---> heaters draw 238x the current, no upset

[trobbins]

Sluckey may have been suggesting to use two 25k trimpots in parallel, so that each 6BQ5 has its own bias adjustment (but also needs additional 25uF filter).

Franco, it may be better to swap 12AX7 heater I with II, as that would make #1 heater more negative (to perhaps make hum less) and make #2 heater less negative (and hence reduce stress on that heater as one triode is in a concertina with perhaps a significantly high cathode voltage).

[sluckey]

Sluckey may have been suggesting to use two 25k trimpots in parallel, so that each 6BQ5 has its own bias adjustment (but also needs additional 25uF filter).

What I posted is what I was suggesting. Simply replace the 10K with a 25K pot. Easy peasy. If you put two 25K pots in parallel without doing anything else, the bias range would be -25v to -13.6v, and that may not be a good range for EL84s. If dual bias pots is desirable, I would suggest adding another 25K pot in series with another 10K resistor. This would allow you to have the same bias range of -25 to -9. But does this little amp really need dual bias? Heck, I would do away with the fixed bias and just convert to cathode bias. 

[kagliostro]

Many Thanks Steve - HotBluePlates - Trobbins

In really a few time I considered more options

... I would do away with the fixed bias and just convert to cathode bias.

Initially I was thinking to a single adjustment like that Sluckey submitted, then go to think it will be useful to have independent controls to use every tube,  also scarcely matched, after that the idea to swap for a Cathode Bias arrived and, at the end, the decision is to use a switch to swap from Fixed Bias to Cathode Bias

Wanting to use a DPDT switch to perform the swap between Fixed Bias and Cathode Bias I'll use only one adjustment on the Fixed Bias (one for both tubes, not one for each tube)



I would like to adopt something like in the Marshall JCM900 (and the previous schematic) where a trimmer is connected as a rheostat and that way also if the wiper lose contact the Bias will be always present





But .... which values are better to be used to have the larger regulation range ?

---

@ HotBluePlates

Good explanation

You wrote 15K5ohm on the Geloso plan, was that a typo ?


@ trobbins

... it may be better to swap 12AX7 heater I with II, as that would make #1 heater more negative (to perhaps make hum less) and make #2 heater less negative (and hence reduce stress on that heater as one triode is in a concertina with perhaps a significantly high cathode voltage).

Didn't considered that, thanks for the advice, do you think it will be a good choice to add elevated reference for heaters here considering also the presence of the Bias supply ?

Franco

[sluckey]

But .... which values are better to be used to have the larger regulation range ?

I assumed the rectifier output to be -25V and used Ohm's Law to calculate for my circuit. You can do the same for your circuit.

[trobbins]

Franco, the concertina triode cathode idle voltage would indicate what idle and peak voltage difference occurs between that cathode and the valve's heater.  For example, assuming 1mA through the concertina, the cathode idle voltage is about +70V, and the peak may get to about +140V.  The heater is at nominal -12 to -24V, so the difference could be circa 100V idle and 160V peak, which is close or at limits depending on actual circuit voltages (and datasheet used).  Swapping heater 1 and II provides a 12V margin to the datasheet limit.

If you changed to cathode bias then the total dc heater circuit could be isolated from 0V (assuming the filter caps can be isolated easily), and that heater circuit connected to an elevated DC voltage (even just connecting the negative end to 0V) so as to reduce any stress on the heater-cathode insulation, but note that the heaters are already effectively DC powered, so any benefit of elevation due to residual hum is likely down in the weeds anyway.

PS.  In case you didn't notice, #1 tube heater gets a more filtered DC than #2 (due to an extra stage of RC ripple reduction), which is why they chose that original location, along with the grounded cathode input stage (and hence grid leak bias) for lowest hum from the very low microphone signal input.

[HotBluePlates]

@ HotBluePlates

You wrote 15K5ohm on the Geloso plan, was that a typo ?

I wrote what I meant to say, but unfortunately I was wrong!   

I saw "15k 5 %" and somehow just ignored the "%" so I saw "15k5."  That extra 500Ω doesn't really change the Ohm's Law calculations in a meaningful way.

[kagliostro]

Thanks Trobbins

I know, no better Hum rejection, it will be only a safety shrewdness

But I'm not still intentioned to swap definitively to a Cathode Bias, only a switch to swap between two different Bias modes


@ HotBluePlates

I saw "15k 5 %" and somehow just ignored the "%" so I saw "15k5."

I was hoping something like that (or a typo), those 0.5K appeared to me like a rabbit pulled out of a wizard's hat  and I was thinking from which obscure math they appeared, may be there were involved the heater resistances values or what else 

When you live in the ignorance all what you don't know assumes a connotation of magic 

That extra 500Ω doesn't really change the Ohm's Law calculations in a meaningful way.

Yes, sure   

Thanks

Franco

[Williamblake]

On a sidenote: Looking at amplificatores is a nice contribution. Keep on keeping up Corto!

[HotBluePlates]

... those 0.5K appeared to me like a rabbit pulled out of a wizard's hat  ...

If I post something that looks wrong, it's probably because I screwed up!  I do it quite a bit.

[PRR]

> Wanting to use a DPDT switch to perform the swap between Fixed Bias and Cathode Bias

[kagliostro]

@ HotBluePlates

If I post something that looks wrong, it's probably because I screwed up!  I do it quite a bit.

My friend, if you say something that I don't understand and seems odd, to me, the first suspect am I, after, may be, you had a little distraction on writing

@ PRR

I see .... I didn't considered feasible to short to ground the Bias supply

That won't affect the supply because of the low current ? 

Franco

[PRR]

This one has the 100k first resistor to drop 350V AC down to like 60V DC. This can be shorted with little increase of heat.

We also have 50VAC taps driving little (470r) to no series resistance. These will smoke if shorted. (We could design them to survive but that adds cost.)

[sluckey]

You must use a DPDT bias switch in that Geloso amp.

[kagliostro]

Better ..... now I understand better, Thanks PRR

and ... Thanks Steve that precisation is appreciated

Franco

 

[PRR]

You must use a DPDT bias switch in that Geloso amp.

Ah, wrong diagram.

[kagliostro]

Ciao PRR

If I read correctly your schematic you are saying that the use of a SPDT is safe because of the negligible current

OK Thanks

BTW Raw Bias Voltage was measured to be -23.3v

Franco

[kagliostro]

This was suggested by our friend Sluckey



Will this work ?



Thanks

Franco

[sluckey]

That will work but you may have to recalculate the voltage spread.

[kagliostro]

To me (I often make big blunders) seems all fine 







Franco