Tubelab Simple SE low B+

[mpri]

Hi all.

I have just completed a build of the Simple SE. After doing some initial resistance tests I put tubes in to do voltage tests and the B+ seems low for what I was expecting.

Here are my PS components using a CLC filter:
Allied 6K7VG 750VCT
35mfd motor run cap at C1
Triad C-14X 6H 150 ohm choke
100mfd Electrolytic cap at C2
50mfd motor run cap optional auxiliary cap for C2
5U4GB rectifier tube
CL-90 inrush current limiter on main fuse/switch side
CL-90 inrush current limiter on PT secondary CT (red/Yel)
Wall voltage is 118V

With this my B+ measures 391V. PSU Designer II shows it should be around 435V. Looks like my system is about 40V less than it could be. I know I can use a 5AR4 rectifier and get about 30V more. However I have installed a bank of switched resistors for tube rolling based on an expected B+ of 450V– 460V with a 5AR4 and I’m thinking that will only get me to about 425V or so.

The amp does work. Just sounds a little underwhelming as the cathode bias resistors are wrong for the B+. I did try another 5U4GB, but it didn’t change much. My questions are does this voltage seem reasonable? Is there something I could do in the PS filter to get more out of it? How do others get 450V?

Any info or suggestions would be much appreciated.

- Mike

[Jebem]

do You mean this amp?
Simple SE

[kevinkr]

When you modeled the supply in PSUD did you use realistic numbers for the winding resistances? Some of the difference may also be due to modeling with with a nominal line of 120V and having slightly to significantly lower line voltage.

Have you measured the supply current with the amplifier operating, and is it close to what you assumed in PSUD?

Modeling gets you close but can be easily be off 5% or more depending on the assumptions you have made, and how well the model matches the components you are actually using.

[Jebem]

I would adjust the tube output bias to the correct value, and this depends on the selected output tube (wich one are you using?), not the rectifier tube. The +B voltage difference is not the main issue, but, as You state, will have a effect on the bias as well.
Check the output tube Ia/Vg curves, find the correct operating point for your +B voltage, and from this You will find the resulting anode current for the choosen Vg voltage; just make sure You are not exceeding the tube maximum power dissipation limit (multiply the Anode voltage by the anode current).

Then get the correct cathode resistor by dividing the Vg voltage by the anode current.

To find out the cathode resistor dissipation, multiply the Vg by the Anode current, and then multiply again by a factor of 4 (to make sure the resistor will nbot overheat).
For example, for a Anode current of 50mA and a Vg of 10Volt, the dissipation will be 50x10 = 500mW; so the resistor should be of 2Watt series.

[jrenkin]

I had a similar problem recently on the Tubelab Se I was building. Turned out I had forgotten to ground C4 (C1 for you) in my point to point vesion of the amp, and so it was a defacto choke input and the voltage was low. May not be your problem, but just a thought that it may be a simple wiring issue.

[mpri]

do You mean this amp?

Yes.

When you modeled the supply in PSUD did you use realistic numbers for the winding resistances?

It's one of those kinda/sorta things. I didn't use the exact resistance since I modeled it after I built it and B+ looked low. But I did take the PSU 330v 31 ohm default and upped it to 375v 45 ohms.

Have you measured the supply current with the amplifier operating, and is it close to what you assumed in PSUD?

No, I will be doing that tonight.


As I said I really modeled it after the fact. I've been looking to build the Simple SE for a while now. I've taken my time getting parts and reading what others have said about their builds. It seems with this setup most people get about 450-460v B+, so that's what I was expecting. When I didn't get it is when I modeled it to see what may have gone wrong.

The one difference I have from others is I used a motor run cap for C1 since 500V 105 degree electrolytic caps are hard to find now. It seems some have used an electrolytic with a lower voltage rating and a slight bit higher MFD.


Also in PSUD as far as I can tell C1 is an electrolytic. Mine is a motor run cap. Maybe there is a difference?


Thanks,
- Mike

[mpri]

Jebem – Yes that would work. However I have a rotary 2 pole 6 position switch with 10 5W resistors soldered on, in parallel to a 1K on board bias resistor to change the bias for different tubes. Currently I have 6L6’s in, but will be using EL-34, KT-88, etc. My effective bias resistance ranges from 470 ohms to 820 ohms. I based the parallel resistors on having about 450V B+ as most others have seen. And so I was hoping to fix the PS instead of changing all the bias resistors.

jrenkin – Thanks for the tip. I used the Tubelab circuit board which should have the ground but I’ll check for a cold solder joint there.

[tubelab.com]

I get 430 volts with a 5AR4 and KT88's cranked to nearly 100 mA each. The same amp makes 440 volts with 6L6GC's at 60 mA. My wall outlet reads 122volts. I can hit 460 volts on the KT88's with solid state diodes but the power transformer gets too hot to touch.

Try a 5AR4, it should get you about 30 to 35 more volts depending on line voltage and bias current.

[mpri]

What bias resistors are you using with your KT88 and 6L6’s to get those mA’s? Low I would think. With a 5AR4 I have seen my B+ as high as only 415 volts with a 6L6GT. With that I can get 64 mA with a 470 ohm bias resistor which gets me 23 watts dissipation. I think I would need a whole lot lower value to get near 100 mA on a KT88. Maybe I’m just thinking too high on my bias resistors. I guess if the voltage is low use lower value resistors. If it’s high use higher value bias resistors. Same results right?

Thanks,
- Mike