kt 120

[martin manning]

This would be my put. Two combinations with Marshall and Fender PT and OT, in both cases the 100W OT is run with double the designed speaker load. The power required is not a problem either way. I think I'd favor the Marshall PT version for the upright-mount PT and more output power. New Magnetic Components transformers would be less than $170.

[es345]

Here is the tool, I might have published an older version already here somewhere.
This is an estimation help for a PP powerstage (not UL type) to estimate

the maximum
- output power
- plate dissipation
- screen dissipation
- average cathode current

Green are the input fields, most of them should be self explaining. Some remarks.
- 100% signal drive means maximum clean power
- R coil is the copper resistance of the primary coil from the center connection to one anode connection. This value has a significant impact on the maximum output power.
- the screen current is estimated as a percentage of the anode current. This estimation works ok up to 100% sgnal drive but not in overdrive. During overdrive the screen is still rising in the area when the anode current is somewhat constant. This is not modelled. So be conservative regarding screen power dissipation. Consequently the maximum average cathode current is a lower estimation when having a signal drive > 100%.

So what you can see if you take the Hammond KX or LX. The limiting element of the maximum power is the maximum allowed cathode current of the KT120. So it is not a perfect match. I don't know your design goals (bass amp, clean amp, amp with abilty of heavy overdrive). Perhaps this little tools helps to analyse some configs.

[Andy Le Blanc]

Another thought supported by experience and the tube simulator above.
Usually when I power test with a resistive load, the efficiency belies the wattage of the opt.

for example... the 50w assumption only really makes 30w of power.

I've been looking at opt and the primary ma. ratings of 50 to 60w devices, per side are twice the expected static DC bias.

so I should only get around 70w of measured power at max for 120w plate dissipation.

Can I get away with a 50 to 60w opt? There are more loading options in that watt range.

The power test is over a resistive load, you literally ramp up a signal voltage,
reading the V across R until the V just starts to drop, thats where you grab your dynamic circuit measurements for maximum-signal figures

I do like over sizing components for reliability but is it really necessary?
How close can you cut it?

[Andy Le Blanc]

I appreciate the modeling, 4k loading doesn't look too bad, I found a jcm45/100 100w opt with a 4k primary, wouldn't have to mismatch an still have a 100w opt, and I can get 4k out a twin opt.

The thing I like about power testing is that it gets you real data, once you find the difference between static and max current figures, it's an open door
to a full description of the circuit.

that marshal PT is full wave bridge?

Is that modelling spread sheet for microsoft office? will it work with open office?

[martin manning]

that marshal PT is full wave bridge?

No, FW. This one: http://www.classictone.net/40-18024.html

This is your JTM45 100W OT w/ 4k Raa and 4-8-16: http://www.classictone.net/40-18067.html

I think these two are it!

[es345]

Is that modelling spread sheet for microsoft office? will it work with open office?

Yes, this is an excel spreadsheed, format is excel 2003. I don't have open office, but this is the old format so I would expect that it will work. Inside there are only simple calculations.

Your question about over or undersizing the OPT.

If your designing a bassamp I would order one which fits to the max power, e.g. 300W with a frequency response of 20Hz-20kHz (- 3db limits) at full power.

If it is a guitar amp, from my experience you can do some compromise. I have another Amp with 2 KT120, about 130W, Raa 4K, Ub 610V. The OPT spec of the ordered OPT was 70W, frequency response 25Hz-54kHz (I didn't request the higher frequency, it was just built in). This one works fine without any problems in the config discribed above - using a guitar. Using my 5 string bass the lower octave is not impressive.

[Andy Le Blanc]

thats a voltage doubler... 350v ct at 290ma... that'll get me 700v and 145ma

I'm aiming for a lower volts and more current... nice tone in there

I know the higher plate voltage will get me more apparent watts with less current and has fidelity benefits, but I like going the other way, I don't want the extra bandwidth.
So many amps are hard toned, my favs have always been softer thicker with more even order harmonic distortion.

appreciate the format info...
I usually go the other way with parts, make it strong enough and then double it.
I don't think I've seriously tried going the other way.

I've drifted quite a bit in overall concept, I got tired of loud fender clones a while back and just kept going
I know I have a few 50w opt around to test the concept with kt88 with more current

[Andy Le Blanc]

ClassicTone # 40-18067, Marshall JTM45/100 Style,100W Output Transformer, 4/8/16 Ohm

I favoring this one too, its affordable.... $99

I still favoring the hammond PT... angela.com (anybody order from them?)
has a 372kx and 372lx for $199

nothing cast in stone

[martin manning]

thats a voltage doubler... 350v ct at 290ma... that'll get me 700v and 145ma

Not really a voltage doubler, it's FWB (my bad) with the CT used to balance the voltage on the reservoir caps. Still gets 470VDC at 290 (a la Marshall 2203), and you could confirm that current rating wired as FWB with Magnetic Components... or did you want to use a vacuum rectifier in FW?

The Hammonds are just too big... I don't see the need for more than 250 mA, and that's derated 25%.

373HX maybe? 700V CT @ 400mA... still overkill and $199.

[M Fowler]

To get the full potential of the KT120 why wouldn't you want your plate voltage higher?

I have used the KT120 pair with Marshall JCM800 PT/OT and it sounded great but no more perceived wattage.

I want to use Edcor 500mA 500-0-500 and similar high wattage Edcor OT.

Mark

[Andy Le Blanc]

oh jeez... Ya I was thinking (ouch) of a couple 5u4 parallel, similar to a 5e6.

I'm trying, over all, to center a conservative max dissipation figure for the tube type
around about 50% of the current rating of the PT.

I know it might seem odd but that's the idea I've been noodling with.
I have a working concept rig on a 200ma PT, its looking do-able so far with the dual purpose of exploring the kt 120
but it defines the part search... 250 to 300 ma static means something in region of 500 to 700 ma.
around about fender voltages after rectification and its ok to be on the low side for volts if I can find the ma.

that 373hx is in the neighborhood, that'll be... 485 to 490ish and land the design max around 60% of the pt rating

[martin manning]

Hammond 378X 400-0-400 @230mA, 5v@3A, 6.3V@6A, 4k a-a load (Marshall JTM 100 OT). FW silicon rectifier gives 536V on the plates, 275 screens, theoretical power 118W. Is that what you're talking about, Mark?

[Andy Le Blanc]

I tried higher plate voltage with kt90, I kept falling back to a moderate plate voltage.
I think it comes down that basic relationship of voltage to current.
push the voltage or push the current, a cathode bias and a lower plate voltage has its own qualities,
a fixed bias and a higher plate voltage has its qualities too.

same load, same operating point, two different beasts...

[Andy Le Blanc]

If I remember, V/I relationships in the powerside have more to do about the opt.
The opt behaves more ideally with less DC current, the bandwidth and frequency response are improved, but as "musical" instrument you can argue the other way.
It becomes less about fidelity and more about harmonic content.

[M Fowler]

I just figured I could get the same results in a guitar amp using 6550 or KT88 if I am going to keep the plate voltage down.

There would be no reason to use the KT120 tubes. They did sound nice and full when used in my Marshall style chassis using MC 100w iron.

The tubes are physically so tall that additonal cab height is needed.

I'm still not sure about using 1k2 primary and would you have to use 120-150w rated OT?

Mark