FBT 500R Power Supply Mods

[dankinzelman]

Hi everyone,

I'm fixing up my third FBT 500R amp. Schematic is attached, and seems to match the amp quite well, with two exceptions (I've drawn the mismatches in pink). The power supply A node has a 5 ohm resistor in between the reservoir cap and each diode, which is not shown on the original schematic. Can anyone explain its function to me? The other difference from the schematic is that all three amps (two of them looked like factory to me) have the PI fed from the B node instead of the C node, and this introduces significant hum to the signal. Ripple at the B node is around 200mV peak-peak, whereas the C node is pretty much ruler-flat.

I've moved the PI onto the C node for the first two amps (which are my own, so I'm willing to take some risks), and they are very quiet now. But now I'm working on a friend's amp. He tours regularly with it, so I am concerned about reliability or potential disadvantages in the medium to long-term. If it sounds so much better on the C node, why did they put it on the B node from the factory? Are there potential disadvantages from moving the PI down to the C node?

Final question: These amps have a fuse receptacle which can be rotated to accommodate different mains voltages. Available settings are 280, 220, 160, 145, 125 and 110. I'm in Italy, so wall voltage is ~220VDC, but my internal voltages are much higher than the schematic if I use the 220 volt setting. I get very close to the voltages shown in the schematic when I set the fuse to the 280 volt slot (ie, internal voltages are lower). I'm concerned about longevity since the EL 503 output tubes are very hard to source. Is keeping things a bit cooler a good plan to prolong tube life? Are there any potential advantages to be had by running higher internal voltages (ie switching to the 220v tap)?

Bonus question: The schematic shows output tube plate voltages of 365V, whereas the OT center tap is speced at 355V. Is this drafting error, or is there some reason why the plate voltages would actually be higher than the center tap?

[kagliostro]

In Italy we don't have any more 220V AC line nowadays line is at 240V so you can experience higher values on the PS nodes than expected

Unfortunately on the TA there isn't a 240V tap so the only option for you is to connect to the 280V if you want to lower the HT

Give a control to the heater voltage with the line connected to the 280V tap and report it here

If the voltage is too low it will be better for you to use a buck transformer to obtain the correct heater and  HT voltages


Note that the connection of the added transformer must be done the right way or you'll obtain a boost effect instead of a buck effect


---


Some time ago I fixed a similar problem in a portable record player (the difference was I there there wasn't a real Transforner and the PS was needed via an Autotransformer) but you can adopt the same solution


See the attached schematics

Franco

[AlNewman]

I would think the 5R resistors are likely used as a fuse more than anything.

The difference between B and C node besides filtering would be plate voltages.  Maybe there would be slightly more clean headroom running off B node, but I think most amps run the PI from the next node after the screens.  I doubt it effects reliability either way.

I think the voltage discrepancy between B+ and the plates is a schematic error. 

[kagliostro]

I forgot to say, to me 5R resistor are there to smooth a bit inrush current

Franco

[pdf64]

Yes, my guess is they're intended to reduce peak current, at power up.
They may also help to reduce switching noise.

[dankinzelman]

Ok thanks for your responses.

My friend's amp has clearly been modified by a tech before it got to me. The work looks quite sloppy to my eye (although apart from the PI hum, it seems to work well). The 5R resistors are not there on this amp. I don't know if they've been removed by the previous tech, or were never there in the first place. Do you recommend I put them back in? If so, what wattage do I need?

So consensus seems to be no problems switching the PI to the C node? Kagliostro do you agree? You seem to know these amps...

I blew up my multimeter the other day so I can't immediately measure heater voltage, but will update when I do.

[kagliostro]

To me de amp was modified to bypass the higher HT problem due to the higher voltage at the line but, to me, this was a short way solution, not an optimal solution

The best solution is to lower the whole HT to the planned voltages, so a buck transformer is the solution

The buck transformer you need is a 240v / 20v transformer (or near 20V, the more close you  can find)

This transformer will be a small one, the fuse on the primary of your amp is 1.6A, so a 240v / 20V @ 2A will be more than enough for your purpose

If you want you can put 2 x 5R resistors on the PS, about the power of the resistors, the bigger they are, the better, without math, I think that 10w will be more than enough for it, I'll prefer an aluminium version, they are smaller than other kind

Franco