Hi,
I have completed my Quasimodo build, and the first three transformers I have tested, with Cx=10nF, Cs=150nF, have shown ideal Rs between 14Ω and 20Ω. Therefore I'm not sure these numbers are correct. My - old - Tektronix scope doesn't show a picture like those posted here before, but rather one peak followed by a few decreasing waves. If I try to minimise them, I always end around 14Ω to 20Ω.
Are these numbers realistic? I'm somewhat suspicious that something is wrong here...
What might be the issue? The scope? Something in the build?
If it's the scope; is there a way to check it? Are they alternatives to using the scope?
The trans I tested were:
1) toroid 230V/500VA 2x40V secondary: Cx=10nF, Cs=150nF, Rs=16Ω
2) toroid 230V/15VA 2x12V secondary: Cx=10bF, Cs=150nF, Rs=18Ω
3) original from Quad 405 from 1977 - 230V/300VA 2x35V secondary: Cx=10nF, Cs=150nF, Rs=16Ω
I have completed my Quasimodo build, and the first three transformers I have tested, with Cx=10nF, Cs=150nF, have shown ideal Rs between 14Ω and 20Ω. Therefore I'm not sure these numbers are correct. My - old - Tektronix scope doesn't show a picture like those posted here before, but rather one peak followed by a few decreasing waves. If I try to minimise them, I always end around 14Ω to 20Ω.
Are these numbers realistic? I'm somewhat suspicious that something is wrong here...
What might be the issue? The scope? Something in the build?
If it's the scope; is there a way to check it? Are they alternatives to using the scope?
The trans I tested were:
1) toroid 230V/500VA 2x40V secondary: Cx=10nF, Cs=150nF, Rs=16Ω
2) toroid 230V/15VA 2x12V secondary: Cx=10bF, Cs=150nF, Rs=18Ω
3) original from Quad 405 from 1977 - 230V/300VA 2x35V secondary: Cx=10nF, Cs=150nF, Rs=16Ω
Worked just like @Dennis Hui said and the boards just arrived 20 minutes ago!
Many thanks forum member rolotube for sending me a free PCB. Finally I have populated the board. Just tried an eBay random transformer and I feel it's way more stable than the "cheapomodo" , thanks Mark for the great project. I can precisely adjust the pot to have the images below, but I'm wondering which one is better tuned?
@peedros I too have extras that I ordered from JLCPCB using Marks files. If @audiosteve cannot I can. Just need compensation for the shipping, I will glady share the board with you.
I just wanted to say thank very much for doing this and writing it up nicely.
I wanted to do a cheaper version (because, only two different transformers in the foreseeable future). I already have a signal generator in my USB scope thingie. So, I reckoned I needed a power MOSFET. And a driver IC. The first driver (if not the only) at my local electronics store (1) had a built-in 555, at which point I gave up and built the whole thing, if only on a breadboard. I used two 9V batteries for power, the IC needing greater than 16V according to specs, and set the oscillation frequency to 100 Hz, 68 kOhm RT and 100 nF CT.
Most of the schematic (just add 4k7 between VBAT and drain and socketed C2 cap between drain and output, plus C3 and RV1 to taste) using IR2153D timer/driver and IRFZ24N FET while disregarding the high-side driver built in (10R gate R) :
Resistor between VBAT and VCC to set operating current (15.6 V Zener diode builtin) to up to 5 mA.
The breadboard setup kind of worked. I would probably have bought a solder-track experimental board if I had to go back to the store again and get more stuff for this.
Highly amusing results, and definitely worth it!
I have two transformers to build into DIY projects, one high voltage toroidal with the max voltage 2x288V (and a bunch of higher current 265V). The 2x288 winding (two secondaries in series) rang like a bell, using 8.2 nF C2 and no damping. The red trace is the Quasimodo output and the blue trace is the LO output of the driver/timer IC. I left the untested secondaries open for the test, but shorted the primary. My thinking (to use a big word) is that the diode bridges are more likely to be conducting at slightly different times, so most of the secondaries are unloaded while the others ring.
The voltage shows ringing at 7.5 kHz which is entirely at the wrong kind of frequency.
whereas with additional C3 of 100 nF and trimmer set to 610 Ohm, it was looking smooth.
Oh, and I tried with a ceramic cap, 2 nF, for C2 and no damping components. The idea being that maybe the loss in the ceramic was enough to dampen the oscillation. It wasn't, with a slightly higher frequency of ringing and four periods of settling.
The other transformer is in the photo above, and is higher power, lower voltage, non toroidal. The oscillation frequency is higher, which is consistent, I guess, there being much fewer windings for 2x33 than for 2x288. Not a lot of attenuation, though. This is with 8.2 nF C2 and nothing else. 125 kHz ringing, is out of the way, I guess.
With C3 100 nF and 44 Ohm RV1, it calmed right down.
So, I am now off to buy snubber components and feeling pretty sure that the result will be as good as I'm likely to get.
Thanks again for the work, everyone.
Best Regards
Jens
(1) Support your local electronics shop, if you have one.
I wanted to do a cheaper version (because, only two different transformers in the foreseeable future). I already have a signal generator in my USB scope thingie. So, I reckoned I needed a power MOSFET. And a driver IC. The first driver (if not the only) at my local electronics store (1) had a built-in 555, at which point I gave up and built the whole thing, if only on a breadboard. I used two 9V batteries for power, the IC needing greater than 16V according to specs, and set the oscillation frequency to 100 Hz, 68 kOhm RT and 100 nF CT.
Most of the schematic (just add 4k7 between VBAT and drain and socketed C2 cap between drain and output, plus C3 and RV1 to taste) using IR2153D timer/driver and IRFZ24N FET while disregarding the high-side driver built in (10R gate R) :
Resistor between VBAT and VCC to set operating current (15.6 V Zener diode builtin) to up to 5 mA.
The breadboard setup kind of worked. I would probably have bought a solder-track experimental board if I had to go back to the store again and get more stuff for this.
Highly amusing results, and definitely worth it!
I have two transformers to build into DIY projects, one high voltage toroidal with the max voltage 2x288V (and a bunch of higher current 265V). The 2x288 winding (two secondaries in series) rang like a bell, using 8.2 nF C2 and no damping. The red trace is the Quasimodo output and the blue trace is the LO output of the driver/timer IC. I left the untested secondaries open for the test, but shorted the primary. My thinking (to use a big word) is that the diode bridges are more likely to be conducting at slightly different times, so most of the secondaries are unloaded while the others ring.
The voltage shows ringing at 7.5 kHz which is entirely at the wrong kind of frequency.
whereas with additional C3 of 100 nF and trimmer set to 610 Ohm, it was looking smooth.
Oh, and I tried with a ceramic cap, 2 nF, for C2 and no damping components. The idea being that maybe the loss in the ceramic was enough to dampen the oscillation. It wasn't, with a slightly higher frequency of ringing and four periods of settling.
The other transformer is in the photo above, and is higher power, lower voltage, non toroidal. The oscillation frequency is higher, which is consistent, I guess, there being much fewer windings for 2x33 than for 2x288. Not a lot of attenuation, though. This is with 8.2 nF C2 and nothing else. 125 kHz ringing, is out of the way, I guess.
With C3 100 nF and 44 Ohm RV1, it calmed right down.
So, I am now off to buy snubber components and feeling pretty sure that the result will be as good as I'm likely to get.
Thanks again for the work, everyone.
Best Regards
Jens
(1) Support your local electronics shop, if you have one.
Oh, and one more thing. I was leaving the unmeasured secondaries open in my measurements above, which is not actually what the nice article (the PDF) says. So, with shorting all the windings not being measured, I get very different results (smaller resistance values needed). Interesting.
Best Regards
Jens
Best Regards
Jens