Scott 299C Bias Circuit 'update'.....suggestions?

[Wharfcreek]

Drew, nothing against the Hayseed option, but I agree with you! If it's 'not' beyond my capability or 'reasonable' effort, then I'll save the $35 every time! Particularly when I've already got the 4 caps needed, and some perf board from which to make the new 'can end cover'.

As to the issue of all the parts vs a 'simpler' less complicated circuit, I'm still up in the air about this. I too feel the natural tendency to go with the manufacturer's 'original' design and just stick with it. BUT....in considering all the other many, many, many amps that I've worked on where the bias supply circuit had only 1 or 2 caps (1 per channel like the Dynaco ST-70 circuit)..... I can't help but wonder just what the harm would be to 'attempt' a part-reduction in this circuit......as long as voltages came out properly. Obviously there's a reason for the way Scott did it.....BUT...... how many things are changed 'these days' due to changes in parts availability....like tubes, cap values, and even major design changes? Could it not be any more simple than assuring that the voltage at the output of the filter section 'to' the four 12AX7 filaments is 46 +/- volts?.......and that the filter section is 'adequate' to keep any hum from filtering in to the input through the negative bias? Seems to me that it 'could' be that simple......and part of me wants to try it!!

TSD

 

[knockbill]

Tom,,
I think you may find that the multiple caps in the heater string are part of the filter circuit,,, the extra resistors along with the multiple caps all help keep it quiet... I recently built a PS for the Fisher 50C clones I'm working on with a series string for the heaters similar to the Scott, and found one stage filter wouldn't kill the hum... But,,, you are curious like me,,, so try it and see how many stages you need!!!!

 

[Wharfcreek]

Hey John,

I guess I'll have to try it and see. Saying that......I have to weigh the aspects of 'trying and failing'...vs just doing it the way it was designed and being done with it!

What to do???..lol

 

[knockbill]

Yep,,, your call,,, I'm guessin they wouldn't have used 4 stages if less would work!

 

[gadget73]

I generally assume that mfg don't put in more parts than they really need to get the job done.

 

[Wharfcreek]

It still seems odd that SOOOOOO many other amps don't have a 4-stage filter system in the bias supply. And, while the DC in this section is used to it's advantage by making quiet some perhaps otherwise 'noisy' filaments if using AC like other amps......... I'm wondering if this filter section is more at targeting any potential noise in the pre-amp filament section vs what it does for the Bias supply?

 

[trainbuftony]

Keep in mind that this supply is providing bias to the grids. Any ripple will be aplied to the grid and amplified. They should be as clean as possible which is probably the reason for the filter design.

 

[Wharfcreek]

Train, I get that.....but how many other amps have a similar 'negative voltage' bias supply that has only one filter cap in it? Granted, in many if not most of those, the addition of having the filament string also involved isn't necessarily part of that design. But, how many 'other' amps don't even have a DC filament supply at all? There's just something about this that is 'odd' to me. It just seems WAY over complicated to me.......such that I just want to understand why the design in the first place. This is really as much about learning and understanding the whole point of the design as it is about trying to 'simplify' my own unit's repair. In fact, perhaps more about the former since I'm already pretty much committed to completing the repair by following the original design. I will have to be using 100uf/100v caps and re-stuffing the can myself, BUT, that's about as close as one can get in today's world of 'parts availability'!!......lol

Anyway.... just a quest for more and greater knowledge!!!

 

[gadget73]

actually because its feeding the grids, it can be fairly dirty. Noise goes in common mode so it rejects.

Dunno, some engineer decided it was needed. You could probably rework it to use less stages but for all the bother of it, why?

 

[Wharfcreek]

The only 'why?' that makes sense to me is simply because the availability of that filter cap is 'nil'!! Finding 100V can caps isn't quite so impossible....even though the uf value seems higher. But.... perhaps, in giving this some thought, and considering the 'availability' of capacitor values back in the '50s and '60s'.....maybe the design concept included having about 300uf in filter capacity in this circuit...... but finding a cap at that value didn't exist back then. Maybe the thing to do is consider what's happening in this circuit 'as is'. Which, and please correct me if I'm wrong.....but the first stage after the first 18 ohm resistor, simply drops voltage very minimally, but also has the 75uf of filter capacity applied. Then another very minimal drop, and another 75uf of filter.....and then same thing twice more! In the end, I'm curious about any voltage 'increase' from the original output of the bias bridge rectifier? As these caps 'charge'..... do they 'build' voltage like that in some other circuits......or does the resistor value retain a constant voltage throughout the circuit? Interesting.........

TSD

 

[trainbuftony]

Good point on the cancelation in the push pull output. Using 4 smaller resistors spreads the heat out more... I'll bet if we had the catalogs they were ordering from complete with pricing, it would be much more obvious. They seem to use this 4x75uf can on several amplifiers.. could it be that they got such a screaming deal on something that they came up with a way to use it? The world may never know. I love the sound I get out of all 3 of my 7591 based scotts. They are extremely quiet when no source is connected with very little audible hum. I tend to want to leave things as they are on these old scotts because I have no faith in my ability to design something better. Maybe someday that will change, but by all means if you have an idea dont let me stop you from trying it out.

 

[gadget73]

yeah it may be more a heat thing. Also possible the transformer's voltage output is just higher than it really needs to be and they did all the voltage drop stages as a result.

 

[Tom Bavis]

It supplies the heaters too, so has to deliver 150 mA . So it needs much more filtering than a supply that just supplies bias, The multiple stages do spread out the heat, and 4 x 75 may have been the biggest available at the time. Yes one or two section caps could have been higher, but the multiple R-C sections filter better than a single larger cap. Modern caps would make it easy to do in two stages, but why go through the effort when the original circuit works well? One thing to keep in mind - the first resistor after the bridge carries more current (it includes DC plus ripple current) so it may run hotter than you'd expect. As mentioned earlier, go higher wattage if you increase the value.

 

[Wharfcreek]

I just don't like 'doing' something because of what I 'don't know!! In this case, as seems to be rather clear, the reason why Scott chose to do this with 4 caps and 4 resistors vs just one or two caps and one or two resistors, that reason may never be truly known! I will be 're-stuffing' the old cap with 4 new 100uf/100V radial caps......and then rebuilding the section with 2 watt metal film resistors, and I'll use a 5 watt resistor after the bridge itself. These 'new' parts should provide plenty of reliability down the road, all being 'over-rated' as compared to the original parts used. But, down the road and once the amp is fully operational again, I plan to 'detach' the connection at the bridge, and make up a circuit using only 2 caps and what ever resistors are necessary to achieve the proper voltages......and see how it works! I guess 'hum' will be the determining factor.

 

[Dave451]

150 mA is labelled on a lot of the schematics. I measured it once and it runs about there, as per above. That always makes me wonder, why so low if it's supplying four 12AX7 filaments? The tube sheet says they draw 150 mA apiece at 12.6 volts. I know Scott is usually running them at lower voltage, but does it make that much difference. Also, I tried to undersize the first dropping resistor until a wise person pointed out (as they do here) that you have to consider AC and DC when sizing it. Sorry if the answer to this question is obvious, but it's been bugging me for a while!

 

[Wharfcreek]

The schematic I'm looking at has no current indicated......but in calculating the current for the 4 12AX7s, I'm unsure of the current in each of the 4 resistors. So, that may be a big part of this as well. I'd have to measure voltage drop across each of them to know what they, in sum, add to the circuit. When done I'll do that as well as the circuit in whole so as to add to the apparent info on this part of the amp. I'm just swagging at it, but looks to me like that circuit may be pulling upwards of 3/4 of an amp, or 750mA if you include the 150mA X 4 on each tube plus what ever each 18 ohm resistor is pulling. That part is still a question as I have no way to measure it until I actually have it up and running again. Maybe I'm missing something here as well.......so my 'swag' may be all wrong!!

 

[Wharfcreek]

Humbug!!! Just looked again at the 4 caps I have here that I was going to use to re-stuff the can, and they're NOT 100/100.....they're 1000uf !! So, I've ordered a lot of 50 of the 100 / 100 and (For 7.95 plus free shipping) I should have them here in another 10 days or so. Murphy's Law!! Not to mention 'Old-timer's Disease!'. Oh well......

 

[gadget73]

That always makes me wonder, why so low if it's supplying four 12AX7 filaments? The tube sheet says they draw 150 mA apiece at 12.6 volts.

In series the voltage adds, but the current does not. In parallel the current adds but the voltage does not. The bias supply itself should use almost no current besides whatever the voltage divider network uses. I wouldn't expect that to be more than about 1ma.

 

[nj pheonix]

Exactly its 150ma at or about 48-50 but youre dividing by 4

 

[Dave451]

Aha, guys--thanks. Back to basic circuit school for me!