Announcement

Collapse
No announcement yet.

running test and simulation on new transformer to step of from 9Vac to 230VAC.

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • running test and simulation on new transformer to step of from 9Vac to 230VAC.

    So I ordered a new Talema 70051K potted toroid for a small project I'm working on. It's a 15VA transformer with dual 9V windings and dual 115V windings. For my purposes, I'm going to use it as a step up transformer using the 9V connected in parallel and 115V in series. The 9V in will also get rectified and filtered for the 12.6V heaters.
    So I connected the transformer for testing the unloaded voltages and winding resistance so I could design and simulate the power supply. I got unexpected results, but after thinking about it, I think every thing is working okay. But I wanted to run it by you to confirm.
    With 9V on the "primary" LV side, I got around 189VAC on the secondary (unloaded). At first I thought there was something wrong, as I expected to see a higher voltage than 230V unloaded. But after looking at the limited transformer specs from the manufacturer, the 9V nominal secondary voltages were specified under a load of 833mA. The unloaded voltage was stated at 11.1V. That works out to be a turns ratio of 21, or just under. If the 11.1V drops just shy of 19% to 9V under load, then applying 9V should correctly provide 230*.081=186V. Close enough.
    Did I do this right, and do I have this correct?
    I did not account for this, but it turns out to be fortuitous in that I really only needed a DC voltage of around 250V.
    If I have a 50% chance of guessing the right answer, I guess wrong 80% of the time.

  • #2
    Sounds about right. Small transformers generally have very poor regulation, 20% is not unusual at all. 189/233 equates to 18%.
    Experience is something you get, just after you really needed it.

    Comment


    • #3
      One other thing I was curious about; The ≈9VAC input will supply the DC Heaters for a pair of 6943 Pentodes and pair of 6111 dual triodes. Both of these tubes operate with 6.3V heaters. So the 6943s will be will be series connected (@ 175mA), and run in parallel to the 6111 heaters wired in series (@ 300mA), for a total 475mA of DC curret.
      How do I calculate the RMS current on the AC side, and how does power factor figure into this?
      Should I run separate simulations for the heater supply, one at the cold filament resistance and one at the nominal operating resistance?
      If I have a 50% chance of guessing the right answer, I guess wrong 80% of the time.

      Comment


      • #4
        What is your concern? If the source is a transformer only the steady-stare RMS current is of interest.

        I'm guessing you use a resistor to get 6.3V from 9VAC? If that is the case the current is 475mA plus whatever the the 9 to 230v transformer load needs.
        Experience is something you get, just after you really needed it.

        Comment


        • #5
          Originally posted by SoulFetish View Post
          The 9V in will also get rectified and filtered for the 12.6V heaters.
          I don't think 9V rms is enough to generate 12.6V DC. I'd run it through PSUD2 to check.

          Comment


          • #6
            If you have a larger transformer with 9VAC secondary, which you want to supply the step-up Tx as well as be rectified for DC heater use, then you could estimate (as a minimum) the VA requirement of the larger transformer by assuming the VA required for the heaters is twice the level of W used by the heaters (as a ball-park), and similarly for the step-up.

            If your larger transformer is not too large, then the current pulses drawn by the rectifier for the heaters will likely 'flat top' the 9VAC waveform, and hence the peak of the stepped up voltage may not be as high as you could expect, and hence influence any rectified B+ level. The 9VAC will also be bouncing around local 0V level by quite a bit (as the DC heater would be tied to 0V), and have significant harmonics, so perhaps only use it for any accessory task with care for noise.

            On a primary side Tx winding, its resistance effectively lowers the internal AC voltage that is being transformed. And on the secondary side, the transformed voltage on a winding then gets lowered by the current being drawn. You may be able to estimate those drops by DCR and rms current measurements if you wanted, although the waveform distortion will muck around with how you interpret any results. So it is sort of 'suck it and see' when all your final load circuitry is in play. The heater path will certainly load down the 9VAC at power on from both the DC caps charging, and the low heater cold resistance, but that should have nearly settled by the time any loading comes on to B+.

            Comment


            • #7
              Originally posted by nickb View Post
              I'm guessing you use a resistor to get 6.3V from 9VAC? If that is the case the current is 475mA plus whatever the the 9 to 230v transformer load needs.
              I could drop the 9VAC to 6.3V, but with 4 tubes (2 drawing 175mA each, and 2 drawing 300mA each) the current would be twice that at 950mA at 6.3V.
              I'm trying to limit the current demand of the auxiliary power supply. If I can achieve a rectified DC voltage great enough to run them in series, then even if the AC current ends up being close to a wash in either case, a DC heater supply theoretically has the benefit of reduced noise for sensitive preamp circuitry.
              If I have a 50% chance of guessing the right answer, I guess wrong 80% of the time.

              Comment


              • #8
                Originally posted by SoulFetish View Post
                I could drop the 9VAC to 6.3V, but with 4 tubes (2 drawing 175mA each, and 2 drawing 300mA each) the current would be twice that at 950mA at 6.3V.
                I'm trying to limit the current demand of the auxiliary power supply. If I can achieve a rectified DC voltage great enough to run them in series, then even if the AC current ends up being close to a wash in either case, a DC heater supply theoretically has the benefit of reduced noise for sensitive preamp circuitry.
                .

                I'd estimate about 10.5VDC if you rectify the 9V supply. So you have to go for 6.3VAC. That's then 0.95A for the heater and say 250mA-ish for the HT supply. We're looking at about 1.5A from the 9VAC supply.
                Experience is something you get, just after you really needed it.

                Comment


                • #9
                  Originally posted by nickb View Post
                  .

                  I'd estimate about 10.5VDC if you rectify the 9V supply. So you have to go for 6.3VAC. That's then 0.95A for the heater and say 250mA-ish for the HT supply. We're looking at about 1.5A from the 9VAC supply.
                  Yeah, I ran a quick simulation and you and Dave are correct anout the heaters (If I use 9VAC in).
                  Typical operation of a 6943 is 8mA+2mA(screen), and 8mA per triode in the 6111s. I may dump on of the 6111s, or run one of them as a cascode to lighten the load. Of course, using different tubes or rethinking the power supply is an option, but I would like to use what I have.
                  If I have a 50% chance of guessing the right answer, I guess wrong 80% of the time.

                  Comment


                  • #10
                    Old Quadraverb power supply, 9VAC @ 2A?
                    Experience is something you get, just after you really needed it.

                    Comment


                    • #11
                      Originally posted by nickb View Post
                      Old Quadraverb power supply, 9VAC @ 2A?
                      I think we have an old Alesis around, but there are a couple other options, I was thinking about this:
                      https://www.elliottelectronicsupply....ply-ac93a.html

                      Or even
                      http://store.ghsstrings.com/power-adaptor-9vac-wall/
                      If I have a 50% chance of guessing the right answer, I guess wrong 80% of the time.

                      Comment


                      • #12
                        Originally posted by SoulFetish View Post
                        I think we have an old Alesis around, but there are a couple other options, I was thinking about this:
                        https://www.elliottelectronicsupply....ply-ac93a.html

                        Or even
                        http://store.ghsstrings.com/power-adaptor-9vac-wall/

                        I had a couple of thoughts.

                        If you use a 2A power supply you'll get something like 10% more i.e 9.9VAC out at a 1A loading. Also, if you use 2A schottky diodes e.g. Vishay SB240-E3/54 with a drop around 0.5V you be pretty close to the desired 12.6V. Def worth a try.
                        Last edited by nickb; 05-29-2019, 12:17 PM.
                        Experience is something you get, just after you really needed it.

                        Comment


                        • #13
                          Originally posted by nickb View Post
                          I had a couple of thoughts.

                          If you use a 2A power supply you'll get something like 10% more i.e 9.9VAC out at a 1A loading. Also, if you use 2A schottky diodes e.g. Vishay SB240-E3/54 with a drop around 0.5V you be pretty close to the desired 12.6V. Def worth a try.
                          I dig what you're sayin' here, and I'll definitely give it a shot. Plus, even if it's not ideal, the 6111 datasheets allow for a +/- 5% for heater voltage and the 6943 allow for a max -12%. So, If I can get a steady 12V+, I'm probably okay.
                          If I have a 50% chance of guessing the right answer, I guess wrong 80% of the time.

                          Comment


                          • #14
                            Originally posted by nickb View Post
                            I had a couple of thoughts.

                            If you use a 2A power supply you'll get something like 10% more i.e 9.9VAC out at a 1A loading. Also, if you use 2A schottky diodes e.g. Vishay SB240-E3/54 with a drop around 0.5V you be pretty close to the desired 12.6V. Def worth a try.
                            I chose to use the SB540 5A schottky's in this series. Max Vf .48V@5A and probably around .35V at 2A it looks like.
                            Anyway, I took some measurements of the wall supply(9VAC/40VA) I got for around $12 and change after scouring the inter-tubes for too long. I ran a simulation and took a screen shot. Hows this look?
                            Also, what kind of ripple % should I aim for in a DC filament supply?

                            Click image for larger version

Name:	DC_Filament_Supply_40VA•9VAC_Wallwart.jpg
Views:	1
Size:	393.7 KB
ID:	853979
                            If I have a 50% chance of guessing the right answer, I guess wrong 80% of the time.

                            Comment


                            • #15
                              Originally posted by SoulFetish View Post
                              I chose to use the SB540 5A schottky's in this series. Max Vf .48V@5A and probably around .35V at 2A it looks like.
                              Anyway, I took some measurements of the wall supply(9VAC/40VA) I got for around $12 and change after scouring the inter-tubes for too long. I ran a simulation and took a screen shot. Hows this look?
                              Also, what kind of ripple % should I aim for in a DC filament supply?

                              [ATTACH=CONFIG]53903[/ATTACH]
                              The ripple is up to you. When you prototype the design you'll find out the hum level from the heaters. You can then adjust the filtering as you see fit based on your cost vs performance requirements.
                              Experience is something you get, just after you really needed it.

                              Comment

                              Working...
                              X