dropping resistors before diodes vs after- tube recto effect

[chopstuck]

I have tried emulating a EZ81 recto by looking up the resistance on Svetlana's site. I ended up using 2 x 220 ohm 3 watt resistors before the diode string and got the same output V+ as I did with a tube recto.

(This is on a small 7 watt push pull 6BM8 based amp).

When I tried using a resistor on the DC side instead, it gave a more compressed saggy sound. Am I right in thinking the resistors effect on the AC side is more pronounced on the extremes of the 60hz waveform than the unfiltered pseudo DC ?

[Andy Le Blanc]

most info I've dug up present that as a way to distribute or buffer the current
going to the rectifier, so you can safely handle more current per diode.
Increasing the resistance before the device drops the ac voltage but
I think you look at it as the total resistance in series .
I think also that placing the resistance after the rectifier and before the first
filter cap it will act more like choke input load, the peak V DC is restricted
so the amp is starved when it needs to respond to a large signal swing.

[chopstuck]

OK then if my two resistors are in parallel on both legs of an ac signal, does that effectively put them in series because they are 180 degrees out of phase on the waveform ?

[Alexo]

If they are in front of the diodes, wouldn't you only be using one resistor at a time? The R would not have any effect on the negative cycle, as the diode already presents it with an effectively infinite resistance.

With an LC input load, you get a very different output voltage at the first filter cap. With the R between diode and filter cap, it does not behave this way. It acts just like a tube rectifier into a C input load.

Honestly, I think you could put the R in either slot for this purpose, except that you would need 2 R's if before the rect and only 1 if after.

[Andy Le Blanc]

Every power supply has some voltage drop when current is drawn from it.
Its called "internal resistance". You divide the voltage drop by the current draw.
This gives you a supply impedance. The lower the impedance the better the regulation.
Putting the resistors on the ac side maintains the impedance that the
rectifier sees at the filter input.

[chopstuck]

OK then, thanks gents !

My goal was to get the V+ drop of the tube recto without taxing the 6.3v heater supply. I had added 2x 12ax7s and eliminated the EZ81 recto. By my math I still have about 300 ma headroom. The PT is way cool to touch too.

[Merlinb]

When I tried using a resistor on the DC side instead, it gave a more compressed saggy sound. Am I right in thinking the resistors effect on the AC side is more pronounced on the extremes of the 60hz waveform than the unfiltered pseudo DC ?

The current which flows in each leg of the transformer is 1/^/2 times the total current that flows from the rectifier (right before the first cap). Therefore, if you have 2 x 220R resistors before the diodes, then to get the same response with one resistor after the rectifier (but before the first cap) will need a value of 220 / ^/2 = 155 ohms. Apart from that change, the response will be *identical* to the previous arrangement with two resistors.

If you had tried using one 220R (I assume that's what you did?) then you will have produced more sag.