100 watt Marshall JMP/JCM power transformer specs & circ

[sebastian]

A couple of things that I don't have understand never well,also if I had already build about #20 100 watt amps for my little artisan job....

Look at this classic schematics:

http://www.drtube.com/schematics/marshall/2203u.gif

A 100 watt Marshall JMP/JCM 2203 amp schematics,the power section use a secondary 350 volt center tapped.There is a bridge rectifier and 2 electrolitics rated at 50+50mF each,so 100mF each,in series,used as a filter for the HT voltage. The center tap is connected in the junction of the 2 electrolitics.
Many people tell this is a voltage doubler???? at my eyes seem to be a classic bridge rectifier with a total load after the bridge of 100mF/2 = 50mF rated at about 2 x maximum voltage of each dual electrolitics,so a total "capacitor" of 50mF /1000 volt.The center tap in my opinion is connected in this point for stabilize the voltage across the caps without using 2 resistor as in the case of the screen grid voltage,after the choke.
The later JCM800 use only a cap of 50+50mF /500 volt after the bridge,and leave the CT not connected.Many people also say that these Marshall have less filtering?? Not true I think,there is 100mF of total capacity VS 50mF of older marshall like plexy,JMP and early JMC800 like the model of the schematics in the link.The sonic difference is maybe why the old model has a little bit B+ voltage (470 under load VS 430-440 under load of the newer models) and why 2 capacitors in series sums the ESR and the feel is more looser than the more "hard" feel of the newer models (with only one caps),due to the effect of the parameters of capacitors over the B+ impedance ecc ecc ecc

Your opinions ??

One last thing.The specs of replacement Marshall power transformer for 100 watt JMP/JCM amplifiers that one can find in internet are all very different and a bit erroneous.Some say that the HT are 350volt center tapped @ 350mA ?? Some 300mA,some 400mA ???
I had made a large number of trasformers for my 100 watt amplifiers and for having ONLY 450 volt under load and for put out about 120-130 watt rms maximum,I have to use a 350 volt 700mA HT secondary!!!
In the late transformer that I have done I use a 360 volt HT secondary at 800mA and I misure with tester classical JMP specs: 475 volt under load and peaks of 175-180 watt RMS at full power.
Some years ago I install a New sensor 100 watt Marshall replacement power TX in a '77 plexy and put in this beast a quartet of 6550 biased each at 45mA .After one our of playing with a THD hotplate at full power the PT were blowned...
What a big quality transformers!!! In an other '83 JCM800 with original Dagnall T4145 PT,I use often 4 6550 biased often @ 65mA each,never blow the power transformer.

[Jack]

a) the 50+50 in PARALLEL makes 100uF. That pair in SERIE with another pair makes HALF the capacitance so 50uF total

b) a bridge recto like that gives 350VCT * 1.41 = 494VAC, so about 480V at the B+. Historically, marshalls used 450 to 500V filter caps which is borderline for a ~500V B+ (especially at startup). So the 2 cans used as a 100uF in serie with a 100uF makes a 50uF filtering at 1000V. On a 50W amp, the plate voltage is low enough to use a single can (or even half a can). Later 100W marshall used the 50W filtering scheme because the quality of the cans increased (or they just crossed their fingers )

The "total" filtering in 50W 3 cans vs 100W 6 cans is the same. The difference is that on the 50W, there is 100uF at the first node and on the 100W, there is 100uF at the phase inverter node. I personnally believe that the tone difference between the 50W and the 100W is due to the 2 extra tube changing the actual class of operation (the 100W is more on the class A side that the 50W). I believe that the perceive tone difference between the 3 cans 100W and the 6 cans 100W is an overblown myth and has more to do with some other "minor" changes in the tone stack and some other places (that weren't on the schematics).

Most PT blows because the HEATER supply is not ok. At least it's much easier to blow the heater tap than the high voltage one.

[mhuss]

What you have in the schematic shown is a seldom seen full wave voltage doubler, basically a both + and - voltage from a single transformer. The "top" 100uF cap is for the - supply, and the "bottom" 100uF can is for the + supply. When you tie the - to ground and stack the two caps, you get effectively double the voltage you would otherwise from a given transformer.

This may have been done simply because they got a deal on a certain existing (lower voltage, high current) PT from Dagnall that month or something.

--mark

[skeezbo]

Is the 2203 the 100 watt Marshall from the '80s with the really high B+ and lots of clean power? I have heard that some of them put out way over 100 watts with 6550 power tubes.
Mercury Magnetics lists power transformers with high B+, but I'm not sure which model they would fit. The voltage doubler power supply sounds like this might be the one.
I was thinking I might try to clone one to use for bass, with preamp changes. Sorry, Marshall guys, if I have a lot of questions; My limited experience has always been with Fenderish clones.
Thanks in advance,
Lou

[paulster]

I'm really not a fan of the term 'voltage doubler' as it isn't per se. It's merely full-wave bridge rectification of the secondary winding.

This actually makes for a more efficient transformer as both halves of the secondary are being used at the same time rather than having one half blocked by a diode for each half-cycle, and is possibly a factor in Marshall's decision as it would theoretically have made for a smaller, cheaper transformer. It also obviates the need for a centre-tapped secondary, although operating in this mode does mean that you have to have a separate bias winding unless you want a capacitively coupled bias circuit.

Marshall used a centre-tap, however, to ensure that due to differing internal resistances on the caps connected in series, that the voltage across each wouldn't stray outside their limits, so whereas we're used to seeing the centre-tap at 0Vdc, in this case it's at 1/2 B+ Vdc. They could just have easily have achieved this with a resistor such as a 220K across each cap. Sometimes they've done both.

I'm still at a loss as to why the 50W amps used a more traditional full-wave rectifier instead, unless it was to save money by providing a bias tap off the HT winding rather than having a separate winding as on the 100W amps.