CurtPalme.com Home Theater Forum

mp20748

One of the next things I wanted to solve was being able to completely turn off my tubes (no filament) when the projector is also turned off. BINGO!

The attached image shows the top of the board showing where I put a FET transistor (isolated) that I used to switch the 24 volt supply to the 6.3 volt filament regulator. In this mod, I'm only adding the FET (IRF520) and two resistors (12K / 2.2K). I put the supply back together checking it out but forgot to take a picture of the bottom...I'll do that later.

This was important to me to not leave filament voltage on the tubes at all times. When the set is "OFF" the voltage is around 3.9 VDC - when the projector is powered up, the voltage is full 6.3 - this mod also eliminates the 3.9 VDC, by removing one lead of a resistor. But with the switch connected where it is, that resistor should not be necessary to remove.

Also, on this LVPS, it is a VDC version that is somewhat different from the original Electrohome design. But the changes should work get the same results when added to that version supply as well..

cmjohnson · Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s

Supposedly, the 3.9 volt standby voltage to the filament is to allow the tube to come up more quickly but it doesn't affect tube life.

I don't know to what that extent is true. It was standard practice with tube TVs for decades.

What would be of greater interest to me would be a true silent standby mode, with no fans running at all when the Marquee
is in standby. I unplug my machine when it's not in use because it runs the fans in standby.

mp20748 · Posted: Wed Sep 13, 2017 3:05 am Post subject:

cmjohnson · Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s

Back in the day, so to speak, a lot of direct view TV sets had an "instant on" feature that ran the filament at reduced voltage when the TV was turned off. I distinctly remembering seeing that faint filament glow on our old Panasonic TV, looking in the vent holes in the back of it, when I was a kid.

After lightning struck our house one night, that TV came on and could not be turned off via the power button. Once the TV was unplugged, and later plugged back in, it worked as normal except the instant on function was no longer working.

Any TV that had the instant on feature ran the filaments at a reduced voltage. And in fact that feature was a discussion subject back when I took my electronics courses in school. Not every set had the feature but it was not exactly uncommon.

mp20748 · Posted: Wed Sep 13, 2017 12:50 pm Post subject:

When the Marquee is in "standby" it's filament is fully 6.3. You can only select standby after the set has been powered on. When turned off, that's when most of the power rails cease and the fans and filament operate at lower voltages.

When in the "standby" mode, all circuits are fully functioning. "standby" mutes the image only on the Marquee if my memory is on point this morning.

When I asked the question, I should have said do you know of a CRT projector that lowers the filament.

I do intend to add a Varister as to limit any surge to the tubes on turn on. There was a technical concern for this, though it appears on Electrohome gave thought to it.

gjaky · Joined: 05 Jun 2010 Posts: 2790 Location: Budapest, Hungary

mp20748 · Posted: Wed Sep 13, 2017 6:06 pm Post subject:

gjaky · Joined: 05 Jun 2010 Posts: 2790 Location: Budapest, Hungary

You could use 12k for both resistor, let's see how that goes.

mp20748 · Posted: Wed Sep 13, 2017 6:55 pm Post subject:

The 12K - 12K combination is PERFECT, when it comes to start (zero), rise (good ramp up), but the voltage maxes at 5.6vdc

gjaky · Joined: 05 Jun 2010 Posts: 2790 Location: Budapest, Hungary

Oh yes, you are using on the positive side... remove the resistor from the gate-source and put a 12-15V zener instead. So 12k from +24V to fet's gate, the 100uF cap between gate-source and the zener parallel with this cap.

mp20748 · Posted: Wed Sep 13, 2017 7:22 pm Post subject:

mp20748 · Posted: Wed Sep 13, 2017 9:09 pm Post subject:

I tried the zener (12) across S - G), cathode to gate. Still quick on.. Sad

gjaky · Joined: 05 Jun 2010 Posts: 2790 Location: Budapest, Hungary

Strange... anyway you can try your original resistors as well (2.2k and 12k) but then you'd need 470uF between the G-S. We should have started with this...

mp20748 · Posted: Wed Sep 13, 2017 10:22 pm Post subject:

OK, I go and make those changes. I was also thinking about increasing the capacitance. I'm also wondering about that FET and should I try a T0-220 transistor instead. But first things first..

gjaky · Joined: 05 Jun 2010 Posts: 2790 Location: Budapest, Hungary

With conventional bipolar transistor you can count with voltage drop between the C-E no less than 0.2V. With MOSFET the voltage drop is given by the Rdson times the current flowing through, in this case 0.6A*0.27R=0.16V. In fact this is also a rather high drop. But with using a different FET, like the IRF3205 (which have an rdson of 0.008 Ohm) the drop could be kept at very-very low level

mp20748 · Posted: Thu Sep 14, 2017 1:51 pm Post subject:

The way i have it now, I'm trying to put the transistor before the 6.3 regulator. I'm now thinking if it would make sense to put it after the 6.3 out from the regulator, in hopes of it ramping from zero to 6.3 for the filaments. What you think?

gjaky · Joined: 05 Jun 2010 Posts: 2790 Location: Budapest, Hungary

Oh, I see. If it is that way probably the best to have your original resistors plus 470uF accross the 12k.

mp20748 · Posted: Sat Sep 16, 2017 11:04 am Post subject:

gjaky · Joined: 05 Jun 2010 Posts: 2790 Location: Budapest, Hungary

Sounds good!
If you want to raise the filament voltage (in order get the proper voltage after the FET), you may want to lower the value of the resistor 4R2 by adding another resistor parallel to it, in the range of 100kOhm, the proper value needs to be experimented of course...

mp20748 · Posted: Sat Sep 16, 2017 7:08 pm Post subject:

That's good to know if I decide to put the voltage point on.

I think I'll look at the ringing issue next...