AC Elevated Heaters . . . .

RickyLee

Well-Known Member
VIP Member
Joined
May 22, 2010
Messages
10,242
Reaction score
3,672
Location
SoCal U.S.A.
I am trying this on my new 2 CH Soldano/MK 2C+ wannabe project. At this point, I am not complaining as the noise and hum is not bad at all. And that is saying a lot as my lead dress in not finished as I have untrimmed power transformer wires still tucked in there from the primary voltage option wires and then I have a 5V tap that I might use to and/or get me 6VDC for my relay coil and possibly putting V1 at 6VDC heaters.

I tapped my faux center tap 100 ohm resistors at the screens node. I have a dual B+ setup of 400V/450V, but will probably stay on the 400V with 6V6 (two in push pull adjustable fixed bias).

Therein brings up my wondering again. I went with only 16uF filtering on my screen, 4H 90mA choke. But my main first filter node is 100uF. So I am wondering if it would have been better to tap into that first filter node?

I will have to go check my work on my divider and cap value on that elevated CT as my memory is getting bad lol. But I have about 42VDC of Ct elevation in low B+ mode 400VDC.

Also, wondering if I would get better bleeder resistor type affect but do not see how that would be possible as I am only downstream from the choke.
 

Gene Ballzz

Well-Known Member
Joined
Feb 19, 2013
Messages
3,539
Reaction score
3,994
Location
Las Vegas, NV
While the 6V6 is certainly capable of operating at 400v, you may find their overdriven distortion to be a bit "harsh" at much above 375v-ish. Are you fixed or cathode biasing them?
Just My Limited Experience,
Gene
 

thetragichero

Well-Known Member
Joined
Jun 6, 2010
Messages
3,427
Reaction score
1,376
Location
upstate ny
add a voltage divider after the last filter node so it's the cleanest dc to "ride on top of". shouldn't need to elevate higher than 60vdc
 

RickyLee

Well-Known Member
VIP Member
Joined
May 22, 2010
Messages
10,242
Reaction score
3,672
Location
SoCal U.S.A.
While the 6V6 is certainly capable of operating at 400v, you may find their overdriven distortion to be a bit "harsh" at much above 375v-ish. Are you fixed or cathode biasing them?
Just My Limited Experience,
Gene

It is all fixed bias, but adjustable with trim pots of course. I considered copying my setup in my old Marshall Studio 15 where I have a switch to go between stock cathode bias and fixed bias. But that difference in volume and feel is actually not much really.

I am running JJ 6V6S right now. They are almost like JJ crammed a 6L6 inside that bottle lol. The data sheet on the JJ's is 500V plate and 450V screen.

I would have loved to have been closer to 365VDC on my low voltage option, but I do not have room for the rectifier tube now. This is all crammed in a Deluxe Reverb chassis. Classic Tone Deluxe Reverb upgrade version xfmrs.

I am going to try some of my old NOS KT66 at the 450V setting, and will probably end up sporting wood on that one. That will probably spoil me on this low voltage option, but it should not be that drastic. This PT and my power supply is capable of 40W, so this is going to be a beastly little 2 channel package amp.
 

RickyLee

Well-Known Member
VIP Member
Joined
May 22, 2010
Messages
10,242
Reaction score
3,672
Location
SoCal U.S.A.
add a voltage divider after the last filter node so it's the cleanest dc to "ride on top of". shouldn't need to elevate higher than 60vdc

The reason I went with the screens node was that was what was recommended to me many years ago by Joey Voltage. It stuck in my head.

When you say last, you mean end of power supply at first 12AX7 V1, correct.

I should have experimented on that really. My work on this elevated CT was done early in the build and it is now deep under other work lol. It would be a hassle to undo it really. It might be easy for me to move it to the main node, but to move it down to the v1 node might be a hassle now. Dam that River, I should have waited and tried it at different locations to see if there is an audible change.

I have another project started and might just do that experimenting on that one. But . . . . that project is going to be a Vintage Modern type layout with either 2nd channel or just a switchable boost stage in there somewhere. So it will not have the same type of possible hum and noise issues as this one here being a MK 2C+/Soldano SLO layout.
 

South Park

Well-Known Member
Joined
Sep 14, 2019
Messages
3,971
Reaction score
3,830
I don’t think the filter cap size matters . It is the DC reference you want for the power tubes heaters . A lot of the SLO clones have this
 

mickeydg5

Well-Known Member
Joined
Sep 30, 2011
Messages
27,146
Reaction score
14,503
Location
united States of America
Moving further down the B+ rail away from the VAC secondary provides more filtering.
Joey recommendation placed it after the second stage of filtering which is more than enough for an elevated filament circuit in this supply circuit I imagine.
 

South Park

Well-Known Member
Joined
Sep 14, 2019
Messages
3,971
Reaction score
3,830
Yes you can connect to the filters at different locations for that circuit . You might get a ground loop
 

RickyLee

Well-Known Member
VIP Member
Joined
May 22, 2010
Messages
10,242
Reaction score
3,672
Location
SoCal U.S.A.
Moving further down the B+ rail away from the VAC secondary provides more filtering.
Joey recommendation placed it after the second stage of filtering which is more than enough for an elevated filament circuit in this supply circuit I imagine.

Who am I to be second guessing Joey's knowledge anyway. It would be just like you giving me some advice and me second guessing what you advised.

:D

I was just thinking that I could have went to a node where I had more filtering. First main node is 100uf, screens is 16uF. Then the phase inverter/reverb is 33uF.

Just wondering if it would have helped a bit more.
 

RickyLee

Well-Known Member
VIP Member
Joined
May 22, 2010
Messages
10,242
Reaction score
3,672
Location
SoCal U.S.A.
I used 2W metal oxides and my divider is a 1MEG over a 150K/47uF cap to ground.



Yes you can connect to the filters at different locations for that circuit . You might get a ground loop

Should not get any ground loops.

I went with a methodical ground scheme, keeping all components/stages grounds within their respective filter node ground, in a flow on a buss wire, from first node to last and then the only ground to the chassis is on the input jack location.

Actually, there is one more ground point and that is the mains ground at the other end of the chassis near where the receptacle comes in.
 

mickeydg5

Well-Known Member
Joined
Sep 30, 2011
Messages
27,146
Reaction score
14,503
Location
united States of America
Practically everything done helps to reduce or rid AC hum in the heaters.
placing heaters in parallel
twisting the wires
using a center tap to ground secondary or balance circuit to ground
elevating the circuit
and even placing cathode bypass capacitors

The fact that you are elevating the the heater circuit places it in a DC plane and attenuates AC current. The additional filtering is icing on the cake.
 

RickyLee

Well-Known Member
VIP Member
Joined
May 22, 2010
Messages
10,242
Reaction score
3,672
Location
SoCal U.S.A.
Practically everything done helps to reduce or rid AC hum in the heaters.
placing heaters in parallel
twisting the wires
using a center tap to ground secondary or balance circuit to ground
elevating the circuit
and even placing cathode bypass capacitors

The fact that you are elevating the the heater circuit places it in a DC plane and attenuates AC current. The additional filtering is icing on the cake.

What I was not expecting was to be able to see a voltage reference from each side/heater leg AC to chassis ground while having the elevated faux center tap. From my 6.8VAC heater tap, I get 3.4VAC each leg to ground. Heater tap windings center tap is not used and I have a faux center tap of 100 ohm resistor each heater leg to that divider point. But I still then have 150K between faux CT point and chassis ground, so I do see there is still ground reference. I had forgot about that last point.

I should have written down my numbers, and will list all those when I get this project dialed in a bit more, but I did see better/lower heater to cathode numbers implementing this elevated heaters. The heater to cathode reading should then be less the elevated reference point, correct? Which in my case is around 42VDC in my low voltage B+ mode of 400VDC. I have a cathode follower in there on V6 and I think the cathode is around 200VDC to ground. So I should then only see around 160V heater to cathode, correct?

I am going to try an old set of my KT66 in high B+ mode 450VDC later on and expect to hear an amazing sound for sure. I was aiming for around 60VDC of elevation in 450VDC B+ mode.

Would an even better elevation method be to strap to the negative bias voltage or any negative DC voltage point somewhere in the amp? Maybe too risky to tap into the bias supply circuit?
 

mickeydg5

Well-Known Member
Joined
Sep 30, 2011
Messages
27,146
Reaction score
14,503
Location
united States of America
@RickyLee
You must try to remember this in your head.
Understand that AC rides the reference line.

If your reference is at 0VDC then a 6.3VAC divides on both sides of 0, +3.15VAC and -3.15VAC do to no VDC reference.

If your reference is at 42VDC then a 6.3VAC divides on both sides of 42VDC, +3.15VAC and -3.15VAC giving totals of 45.15V and 38.85V.
BUT your multimeter can only measure either VAC or VDC at one time.
So if you measure VAC you see the +/-3.15 volts. If you measure VDC you see the 42 volts. You have to use your brain and do the math.
NOW if you had an oscilloscope you could measure and see it all at once. The image would show the 6.3VAC riding at the 42VDC level on its monitor.

And yes, 200-40=160. :)
 

RickyLee

Well-Known Member
VIP Member
Joined
May 22, 2010
Messages
10,242
Reaction score
3,672
Location
SoCal U.S.A.
I been thinking about the orientation of bright caps that are on switches then go to the gain pot. I am wondering if there is a preferred way to install the cap in relation to noise being possibly picked up. Can you see either the cap or the wire from switch being a culprit to picking up noise/RF depending on which side of the switch I put it? Or is this a case of negligible affect and 6 of one half dozen of another? lol

If cap was on out side of switch and attached to pot wiper, then if cap is acting like antenna, at least some low end frequencies are blocked where a wire is picking up/passing all frequencies.

I have noticed some odd issues on my last build with a switch that was close to my gain pot. I found when I moved that switch away, signal and tone changed for the better.
 

mickeydg5

Well-Known Member
Joined
Sep 30, 2011
Messages
27,146
Reaction score
14,503
Location
united States of America
I have always checked capacitors in both directions even if it had shield markings, just to know and make sure.

If required or placed in a critical signal path especially in areas of high impedance circuits use shielded cable/wiring.
 

alpha al

Active Member
Joined
Oct 19, 2017
Messages
101
Reaction score
146
Location
Gastonia, NC
add a voltage divider after the last filter node so it's the cleanest dc to "ride on top of". shouldn't need to elevate higher than 60vdc
A lot of vintage amps that were cathode biased used the 25-30V cathode voltage from the power tubes to elevate the heater supply with DC.

Some even used the preamp tube's heaters as part of the power tubes cathode resistor.
 

RickyLee

Well-Known Member
VIP Member
Joined
May 22, 2010
Messages
10,242
Reaction score
3,672
Location
SoCal U.S.A.
OK, still on the subject of lead dress and keeping hum and interference to a minimum.

I have a 5V tap and the wires coming out of the power xfmr are quite long. I will use this tap possibly to rectify into 6V for my relay coils and I am considering converting at least V1 heaters to this DC 6V as I need a bit more heater current headroom to run my KT66 tubes later on. First half of v1 is for both channels. So CH1 uses V1 V2 V3 and CH uses V1A, and V5 V6. V5 is reverb and V7 is phase inverter. So with that, I might go V1, V2 and V5 all DC heaters from that tap.

Back to lead dress. I have the long 5V tap wires twisted and taped off and laying there. I noticed when messing with the amp that quite a bit of hum came in when those 5V wires got near the phase inverter section of my board. And this new 5VAC to 6VDC tap will need to travel over half way across the chassis towards the input. I am then wondering if it would be better for me to cut those 5V wires short near the power xfmr and convert to DC somewhere near the PT, and then run my now DC power wires across the chassis or just leave the PT wires and run them to where the DC rectifier circuit and relays will be?

That is: Will a length run of AC wires subject noise worse than a length run of DC wires? lol

I am sure a few of you builders has to have encountered this one before . . . .
 


Top