Heater thinker

[Viitalahde]

I got a question.

Have 4 PCC189's to parallel, and I also have a need for a DC rail of about 40 volts.

To reduce the number of transformers, I thought of having a 230vac:30vac transformer and run the 4x PCC189 heaters (=7,6v) at 30VAC, heaters in series & parallel. From the same 30VAC rail,we'd rectify the ~40VDC, and also take the plate voltage through a reverse-connected transformer (18vac:230vac or so).

The question is: Is the bridge rectifier going to inject crap into the heaters?

 

[shabtek]

my guess is that if the dc is well filtered for the current it won't be a problem.
is it wrong to see the DC as a low-z source feeding a low-z load--and any RC filtering action as increasing the source-z of any rubbish on the rail?
that makes sense to me.
but any hash that gets through to the filaments can get picked up by sensitive nodes in proximity.
hope that clarifies things

 

[PRR]

> From the same 30VAC rail,... take the plate voltage through a reverse-connected transformer (18vac:230vac or so).

You can't feed 30VAC into a winding made for 18VAC.

Well, you "can", but it will smoke.

230:30, 30:230 will give over 200VAC and at least 300V DC... isn't that enough?

 

[Viitalahde]

Thanks.

I just quessed the 18vac:230vac transformer as I went on, and obviously it was a brain fart. Of course it's going to be a lower turns ratio for the B+ voltage, which is not going to be that high. I'm aiming for about 130vdc-140vdc regulated, so the raw DC needs are somewhere in the ballpark of 130vac.

Oh, forgot to mention that the circuit is (obviously) a push-pull stage for a vari-mu I'm going to toy around with, so it's capable of rejecting crap.

 

[shabtek]

are you feeding the heaters D.C.?
you do want to rectify the 'interstage' (step-down 2ndary/step-up primary) to get the 40vdc and rectify the step-back-up for da B+, correct?

also yer tubes in series-parallel need 15 volts, in series string 30V

 

[Viitalahde]

Nope, I was thinking of AC heaters, picked right before the 40dc rectifier. This situation calls for an ASCII schematic!

-

                           PCC189-1    PCC189-2
                        o---/\--+--/\---o--/\--+--/\---o   (All dual heaters in series, two
                        |  PCC189-3    PCC189-4 |    tubes in series and two in parallel)
                        o---/\--+--/\---o--/\--+--/\---o    
                        |                                  |
                        |      +-------------------------+
                        |       |
                        |       |             (bridge rectifier)
                        |       o--------------------/ \
                        |       |                   /    \
                        |       |           o-----+     +---------o--------> ~40vdc    
                        |       |           |       \    /            | +
                       o-------|---------------------\ /             =
                        |       |           |                         |  10,000uF/63v
                        |       |          o------------------------+
                        |       |           |
                        |       |         GND
                        |       |
------------3 || 3----o-----------------------3 || 3----------------
            3 || 3             |               3 || 3
            3 || 3             |               3 || 3 
230vac  3 || 3  30vac   |               3 || 3    ~ 130-140vac (for regulated 140v B+)
            3 || 3             |               3 || 3
            3 || 3             |               3 || 3
------------3 || 3-------------o--------------3 || 3---------------

  TX1: 230:30                         TX2: 50:230 (or 25+25:230)

A quick calculation ended me up with a 25+25:230 transformer turned backwards for the B+.

Now that I look at the schematic and the PCC189 datasheet (http://www.mif.pg.gda.pl/homepages/frank/sheets/010/p/PCC189.pdf), I see Shabtek is right. I should indeed have all the heaters in series..
Right?

So it's 300mA each tube.. 1,2A for 4 PCC189's?

 

[thermionic]

This may be a silly question, but could you get enough current from a voltage multiplier arrangement to feed your B+? I've seen this done for 4 plate supplies before, from a 30v secondary (200v B+). However, 2 of the valves were miniature types, so their B+ current would be below that of your PCC189

Also, when I did see the above arrangement (it also had series heaters), the designer added shunts around the miniature valves' heaters so as not to overload them from the draw of the bigger valves. I wonder if shunts are a good thing in any series heater arrangement; from what I understand, heater demands can vary from valve to valve, and a shunt will prevent stress to the less greedy heater.

If you're nervy about the bridge introducing hash into the heater line, how about adding a network of snubber caps, say 10nF between each pin of the bridge?

Also, could you add some bypass caps at the anode pin of each valve?


Justin

edit - thinking about it, the cost of a handful of diodes will be well below that of a second TX. How about having a pair of voltage multipliers, with each one feeding 2 valves?

 

[shabtek]

300mA all day--in series

 

[Viitalahde]

Whoa, of course it is 300mA for the whole series of heaters. Hehe. :grin:

Such a small current might make it reasonable to go for DC heaters after all. I've understood small variations in the line voltage can affect the heaters enough to make the transconductance change a little. Not a big one (and line voltage is pretty stable here) but I'd like this thing to sound the same the next day.

This in turn defeats my worries of anything being injected in the heater lines.. Case closed?