SimpleSE PSUD II B+

[rknize]

1) I don't think it affects the simulation.

2) You probably don't need to mess with these too much. Basically, the goal of those calculators is to try to estimate efficiency (losses in the windings and the core) and regulation of the transformer (how much does the output sag under load). Keep in mind that the voltage at your outlet can easily vary +/- 10% or more, so spending a lot of time getting the model of the transformer right doesn't really pay off. Since you don't have the transformer in your hand, you will have trouble filling-in those numbers anyway.

3) Higher ESR will limit the amount of current that can flow in and out of the capacitor. In terms of a power supply filter, it's like adding a series resistor which will increase ripple. I suggest that you play around with it and see for yourself.

4) Yup, go ahead and reverse them. Take a look at the voltages at C1 and C2. You will see why you will need 500V caps. IIRC, the 3.25mA comes from the bleed resistor on the B+ rail. Then you have 10mA x2 for the 12AT7 plates plus the output tubes. So 175mA is probably a good number for you.

[rknize]

If you want to learn more about what is really going on in this circuit (especially about your #2 and #3 questions), I highly recommend you pick up a book. Morgan Jones' "Valve Amplifiers" is very highly regarded in these circles as it covers these topics from top to bottom. It covers the most basic electronics fundamentals and also delves deep into tube circuit design. So it is great for when you are starting out and will still be useful later on when you get into it more deeply.

[tunertom]

Brad:

Bear in mind that the 6V6 tube has a maximum voltage spec of about 315V for the plate and 285V for the screen, so B+ needs to be lowered to use this tube. One way to do this is to set up your build so that C1 can be easily removed from the circuit. This changes the power supply configuration from having a capacitor load on the rectifier to choke load. Try removing C1 from the PSUD model - the voltage drops down to about 350V@100mA with 5AR4, and 300V@100mA with 5U4. The 5U4 doesn't have the slow warm-up feature of the 5AR4, so even more reason to put in the recommended thermistors.

If you substitute an off-board motor run cap for C1 (say 45uF 440VAC), you could devise some method of safely disconnecting the wires to the capacitor to take it out of the circuit. Voltages are over 500VDC, so exercise due diligence...

Lowering the B+ for the 6V6 of course lowers the voltage feeding the CCS for the 12AT7 as well, so devise a means to short across series resistors R14 and R24 to effectively take them out of the circuit.

After doing all this, you get about 1 watt of supposed 6V6 SE magic (I haven't tried it myself...)

Tom

P.S. My civil engineering curriculum-required EE classes from 40+ years ago left a lot to be desired, so I heartily second Russ's recommendation: I bought Morgan Jones's 2 latest books from Amazon.

[rknize]

I didn't even notice that 6V6 was in your list. Yes, some cleverness is required along the lines of that Tom suggests to get the SSE to be that universal. You'll want a quality choke instead of a cheapo. Disconnecting C1 makes the PSU a pure LC filter. That puts most of the diode switching transients on the choke which will tend to make it buzz. Having some headroom in the mA rating of the choke should help to reduce the buzzing.

I don't have any experience with Edcor's chokes, but I suspect they are decent. I've found Hammond's big chokes to be a mixed bag. Some buzzed loudly while some were quiet. I use mostly CLC filters nowadays, so choke quality is less important.

[squiffiness]

Originally Posted by tunertom

After doing all this, you get about 1 watt of supposed 6V6 SE magic (I haven't tried it myself...)

Tom

Tom,

I think you have effectively talked me out of the 6V6. However, I will use your information to better my understanding of the LC, versus CLC filter.

A thought from the 5AR4 datasheet and PSUDII sims. If you stuff in some resistance at the secondary winding as shown in the schematic pages of the 5AR4, remove C1 and swap in a 10H choke (ESR 75 Ohms), the B+ sits around 260 volts...interesting. Perhaps something to consider after all.

Originally Posted by tunertom

P.S. My civil engineering curriculum-required EE classes from 40+ years ago left a lot to be desired, so I heartily second Russ's recommendation: I bought Morgan Jones's 2 latest books from Amazon.

I actually bought an eBook version last night after Russ Knize suggested in an earlier post. Mr. Jones seems a little down on SE amps in general, but the preamble took me back to my second semester of mechanical engineering school in 1987 when we had EE102 (Electronic Circuit Theory) thrown at us.

Fortunately, my grandparents called tubes "valves" so no adjustment for me.

Originally Posted by rknize

I didn't even notice that 6V6 was in your list.

Russ,

I appreciate all the efforts to assist. Realistically, the 6V6 wasn't on my "must have" list...just wanted to pursue the most flexible build out possible to experiment with this kit longer term since I am on a shoestring budget for a few key reasons and it may be the only thing I can play with for awhile.

Thanks to your suggestion, I can better understand the implications of going from CLC to LC and have already run a sim or two.

BTW -- to all in the TubeLab forum, THIS is why I love the DIY community and homebrew software folks...it's the comraderie, sharing of experience and coaching.

My PCB should be arriving soon. So, stay tuned as I build.

Brad

[squiffiness]

Originally Posted by tunertom

If you substitute an off-board motor run cap for C1 (say 45uF 440VAC), you could devise some method of safely disconnecting the wires to the capacitor to take it out of the circuit. Voltages are over 500VDC, so exercise due diligence...

Were I to go this route, I would likely device some way of switching in a higher power bleed resistor to drain a large cap.

Yet another reason to perhaps loosen my grip on being able to run the 6V6.

I think I might get more joy from being sure I can run the KT88's easily by using RKnize's scheme of a homemad decade (in his case a dodecade ) box of a more humble form (say a trio) to allow a swap of the three main tube types on TubeLab's sims page.

Appreciate all your info though...tremendous help as I get ready to build.

[rknize]

I didn't pursue the 6V6 route since I already had a TSE PCB.

Don't worry about MJ's anti-SE stance. His points are valid ones, but they are irrelevant in the eyes of most SE amp builders. Since the input/driver stages of most PP amps are SE anyway, most of the information still applies. The missing parts can be gleaned from here and elsewhere.

[squiffiness]

To all,

Just finished some DuncanAmps PSUDII Sims for a SimpleSE.

I am using the Edcor XPWR002-120 (360-0-360V up to 220mA) as a basis for the high voltage supply through a 5AR4 and CLC follower.

Interested in some feedback from the group on what I see the sims telling me for driver stage (2x12AT7 @10mA per) plus the various output tubes (EL34's, 6L6GC's and KT88's). Using TubeLab's simulations from the "Tubes and Transformers" page.

What I am putting in for a load after the second cap in the power supply CLC filter (as I1):

1) 2xEL34 (estimate 75mA Ik per tube) - 175 mA estimated load - B+ = 445V
2) 2x6L6GC (estimate 80mA Ik per tube) - 180 mA est'd load - B+ = 438V
3) 2xKT88 (estimate 95 mA Ik per tube) - 210 mA est'd load - B+ = 425V

If I am along the right direction, that allows me (thanks to RKnize's and BoyWonder's help) to select R17/27 requirements to optimize cathode current (Ik).

The Edcor I selected is about $20 cheaper than the XPWR035 that many have selected for their SimpleSE, which is what caught my eye early on. The HT voltage is a little lower, but this does not appear to matter too much.

Thanks

[rknize]

Your numbers all sound about right. Remember that you can easily go +/- 10-20% due to line voltage and other variations, so don't get too hung up on the exact numbers. In other words when you come up with your set of resistor values, add one extra setting on both sides. That way you can shift your final values one way or the other, depending on how your initial testing goes.

[boywonder]

Originally Posted by rknize

Your numbers all sound about right. Remember that you can easily go +/- 10-20% due to line voltage and other variations, so don't get too hung up on the exact numbers. In other words when you come up with your set of resistor values, add one extra setting on both sides. That way you can shift your final values one way or the other, depending on how your initial testing goes.

Looks good to me too (although I'm no expert).....tubes are very forgiving as far as voltage and current go, as long as you don't push the envelope too hard......