Rocket rectifier tubes and Merlin's Rocket mod

[Jackie Treehorn]

Hey, Martin. Whatever it was, it worked and strikes me as a pretty good solution. Look at the Dual Rectifier, with 110 microfarad on a 5u4 rated for 40 microfarad. Just double them up. That's a solution done on a lot of amps. I've never tried the real Mullard, but I would guess that a weak JJ with a very robust hexfred might sound closer to a strong Mullard than the weak jj on its own or an all solid state rectifier.

Adding resistance seems more of an approach to turn the gz34 into a 5u4.

[martin manning]

Sim results show peak current for a hot switch-on is ~2A. Using 40uF vs 80uF cuts down the highest peak significantly, but doesn't mean a whole lot to the size of the peaks after the initial power on, -4% without added resistance, -1% with. Adding series resistance reduces initial and steady state peaks ~15-20%, but lowers B+ 6%.

Max steady-state peak current per plate is 750 mA (Amperex) 825 mA (GE), and GE gives max transient peak current 3.7 A for 200 ms. Looks like the peak current steady state is the one that is limiting with the 80uF Cres and no added resistance. Backing up with SS diodes or avoiding hot switching won't help in this case. Smaller Cres helps some, but adding series resistance is the cure.

Edit: There was a bad number in there...

[Mark]

Hi Merlin thanks for your response, good to have you here. I'm wondering if using the 300 - 0 - 300 tapping and using larger R limit resistors (100 ohms perhaps) appears to be the wisest choice. That way the B+ is preserved and the rectifier is given some protection.

https://tubeamparchive.com/files/taylor ... 08_758.pdf

I had thought of putting one 40uF cap (of the 80uF cap) on either side of the standby switch to counteract the flyback from the choke if the standby switch were used (so much of this power supply seems like a really bad idea at the moment).

Thanks for putting the figures into PSim for me Martin, that was quite helpful.

What are your thoughts in lieu of the PSim results?

I'd even considered a Weber Copper cap.

http://www.webervst.com/ccap.html

[martin manning]

I added some more info after fixing my error. Re the copper cap, it would be an interesting experiment comparing it back-to-back with a vacuum rectifier. You could make up your own version, too. All you need is a couple of SS diodes and a resistor in series with each one to get back to the same B+. Webber's WZ34 listing says 17V drop at 225 mA, and I can find that point on the 5AR4 data sheet curve. 16.3/0.225 = 72R.

[Mark]

Martin thanks for the corrections, so the best valve rectifier option is to add the resistance in or move to solid state and everyone is happy.

I had my suspicions about this sort of thing some months when fixing a Valvetrain amp, nice enough amp but the 5Y3 had to endure a 47uF first filter. This blew the brains out of the 5Y3 and I lowered the first filter cap so the NOS 5Y3 could live a long and happy life.

http://www.londonpower.com/standby-switch#intro