ToneMerc wrote:
All replacement MM PT will run warm with just four EL34's as its only around 5A, smaller ones are slightly less and I even that Hd130 replacement would run very hot with four Kt88's,
If you are truly set on running KT88's you are going to need a much bigger PT. What is the size of your PT cutout? There's only so much wire you can physically wind on a frame. The lamination stack can be made taller but there will be a point where a endbell will not fot over the core. Keep in mind are talking over 1A of HT current and at least 8A of filament current plus bias and another 5V winding, this takes up room.
Dedicated bias winding will have better stability, being totally independant of any draw from the HT winding.
Here's the next question, what about at least a 150-175W output transformer which to matches the KT88 load in this configuration?
Here's the general guide of how you want to order a basic quad EL34 version that fits in a Twin Reverb or JTM factory cutout. Unlike the factory MM version with the underrated filament and CT bias winding, this runs cool.
120V primary
268-0 @ 1A
35-0 @ 300mA
6.3V @ 7A
TM
I'm not set on KT88s but this build will run 2, not 4 power tubes. I can certainly live with EL34s. But the more I think about this the more I realize that I don't understand why the current needs to be so much higher than 400ma in this case. I am planning to run 2 KT88s and 4 preamp tubes with ~700-750 on the plates and 350-400 on the screens. Since I ended up (however mistakenly ) with a 270-0-270 400ma HT transformer, if it I run it with a full wave bridge non center tapped I'm not seeing how the tubes could draw more than 400ma. If I was voltage doubling a 270-0 PT with 400ma HT I could understand why that could be a problem. What am I missing?
ToneMerc wrote:
All replacement MM PT will run warm with just four EL34's as its only around 5A, smaller ones are slightly less and I even that Hd130 replacement would run very hot with four Kt88's,
If you are truly set on running KT88's you are going to need a much bigger PT. What is the size of your PT cutout? There's only so much wire you can physically wind on a frame. The lamination stack can be made taller but there will be a point where a endbell will not fot over the core. Keep in mind are talking over 1A of HT current and at least 8A of filament current plus bias and another 5V winding, this takes up room.
Dedicated bias winding will have better stability, being totally independant of any draw from the HT winding.
Here's the next question, what about at least a 150-175W output transformer which to matches the KT88 load in this configuration?
Here's the general guide of how you want to order a basic quad EL34 version that fits in a Twin Reverb or JTM factory cutout. Unlike the factory MM version with the underrated filament and CT bias winding, this runs cool.
120V primary
268-0 @ 1A
35-0 @ 300mA
6.3V @ 7A
TM
I'm not set on KT88s but this build will run 2, not 4 power tubes. I can certainly live with EL34s. But the more I think about this the more I realize that I don't understand why the current needs to be so much higher than 400ma in this case. I am planning to run 2 KT88s and 4 preamp tubes with ~700-750 on the plates and 350-400 on the screens. Since I ended up (however mistakenly ) with a 270-0-270 400ma HT transformer, if it I run it with a full wave bridge non center tapped I'm not seeing how the tubes could draw more than 400ma. If I was voltage doubling a 270-0 PT with 400ma HT I could understand why that could be a problem. What am I missing?
BTW the transformer has a 6A filament supply
You are right, you don't need as much current with only two OT tubes, nevertheless how are you going to drop 350 volts to feed the screens if you are not using a doubler scheme?
Not sure what you are using for an OT, but the factory load for two EL34 @ 700v/350v is 11K at 8 ohms.
matt h wrote:Just remember to check what your PT current rating actually means. AC current? Or DC recto'd current? (if so, it's likely stated as a FW). When you go to a FWB (vs FW) set up, you're not going to get the full FW DC current rating, but about 60% of it. Thus, your "400mA" is only good for 240mA-ish. (or rather, that's the amount of current equivalent to the same point if it were FW, so drawing more than that will cause PT sag).
In other words, "watts is watts" and there's no free lunch.
I didn't specify what kind of rectification I meant to use when I asked for 400ma so I'm sure Heyboer assumed FW. So taking your 240ma figure, I could be OK. I'm not sure how to properly calculate power tube current draw at 700+ volts but for a point of reference the Dreamtone Current Draw calculator shows 193ma draw for 2-KT88 and 4 preamp tubes @ 582vdc. the calculator doesn't go higher than that but current draw should be somewhat less at higher voltages. If I used EL34s instead of KT88s the draw would be even less.
Last edited by lord preset on Tue Jun 10, 2014 6:10 am, edited 1 time in total.
ToneMerc wrote:
All replacement MM PT will run warm with just four EL34's as its only around 5A, smaller ones are slightly less and I even that Hd130 replacement would run very hot with four Kt88's,
If you are truly set on running KT88's you are going to need a much bigger PT. What is the size of your PT cutout? There's only so much wire you can physically wind on a frame. The lamination stack can be made taller but there will be a point where a endbell will not fot over the core. Keep in mind are talking over 1A of HT current and at least 8A of filament current plus bias and another 5V winding, this takes up room.
Dedicated bias winding will have better stability, being totally independant of any draw from the HT winding.
Here's the next question, what about at least a 150-175W output transformer which to matches the KT88 load in this configuration?
Here's the general guide of how you want to order a basic quad EL34 version that fits in a Twin Reverb or JTM factory cutout. Unlike the factory MM version with the underrated filament and CT bias winding, this runs cool.
120V primary
268-0 @ 1A
35-0 @ 300mA
6.3V @ 7A
TM
You are right, you don't need as much current with only two OT tubes, nevertheless how are you going to drop 350 volts to feed the screens if you are not using a doubler scheme?
Not sure what you are using for an OT, but the factory load for two EL34 @ 700v/350v is 11K at 8 ohms.
TM
The OT came from a 100w KMD that was a close clone of the MM RD 100 running 2-6l6GC so it should be OK
As for dropping the voltage, I don't know yet. One mistake at a time I say. That may be the deal breaker but I need to thrash around a bit more.
lord preset wrote:
You are right, you don't need as much current with only two OT tubes, nevertheless how are you going to drop 350 volts to feed the screens if you are not using a doubler scheme?
Not sure what you are using for an OT, but the factory load for two EL34 @ 700v/350v is 11K at 8 ohms.
TM
The OT came from a 100w KMD that was a close clone of the MM RD 100 running 2-6l6GC so it should be OK
As for dropping the voltage, I don't know yet. One mistake at a time I say. That may be the deal breaker but I need to thrash around a bit more.
You are making this 50x harder than what it needs to be, just use the corect transformer.
ToneMerc wrote:
You are making this 50x harder than what it needs to be, just use the correct transformer.
TM
That is completely true. But my goal is not just to make an amplifier ( I have too many anyway) but more to learn how amplifiers work. That and to provide comic relief to TAG. Every time I try to figure out how to work around a screw up I learn things even if the work arounds don't pan out. In this case I am reading up on SS rectifier types, B+ reduction methods, etc, none of which may lead to anything but at least I will know more next time.
I just talked to Heyboer and they confirmed that if I just use one of the 270 secondary taps (i.e. 270-0) that the full 400ma current will be available.
This means that I should be able to connect the 270-0 taps to the voltage doubling supply I already have and it should work. I still don't get a separate bias secondary and the current rating is a bit tight for what I am doing but it's definitely worth trying, at least with EL34s. Of course the PT is considerably larger than it needed to be because of the CT design but it's already paid for and bolted to the chassis soo....
Resurrecting this thread for an update. Back in June I went ahead and installed the new Heyboer PT using one of the 270 secondary taps. Voltages seemed OK , light bulb test OK so I tried to fire the thing up with tubes. Voltages still as expected, but the amp made a very loud humming noise. Lifted NFB - no change. With the power tubes out I could pick up the bad hum in the preamp, so I assumed I had made a dumb-ass wiring mistake. Many hours later I had had enough - put the amp aside and went on to complete to other builds thinking I would get back to it later with fresh eyes.
Fast forward to October, I felt guilty about abandoning the forlorn Hiwatt and went back in. I did find a wiring error in the tone stack but that was not the problem. However I did find something. I had been trying to trace where the noise was coming into the circuit using a simple DIY rig - injecting a test tone with my iPhone and listening with a DIY battery speaker probe. I was getting the hum on the plates of V1-V3 and on V3 (the PI) even with V1 & V2 removed. It finally occurred to me that there was noise coming from the power supply. Disconnecting the PS from the preamp board I found that the all the nodes B1 - B5 had the nasty hum. Seemed wrong to me but not having ever tried to "listen" to a power supply before I didn't know if this was a valid test so I tried the same thing on a Small Special build I just completed. Hum on B1 and B2 but the preamp nodes were quiet.
So could this be as simple as a bad cap or ..?
BTW the Hiwatt PS is a high voltage design more or less as attached, except I added a zener string before B3 to limit start up voltage to 450v.
You do not have the required permissions to view the files attached to this post.
The choke is in play, too. Try a big resistor of appropriate Ohm value in place of the choke. I'm having a hard time thinking it is the caps, especially if new caps. It is not unheard of for caps to be bad out of the box, but I am thinking chances are small. Can caps actually make noise? Seems odd.
Phil_S wrote:The choke is in play, too. Try a big resistor of appropriate Ohm value in place of the choke. I'm having a hard time thinking it is the caps, especially if new caps. It is not unheard of for caps to be bad out of the box, but I am thinking chances are small. Can caps actually make noise? Seems odd.
Sorry I failed to mention this is already a resistor in place of a choke. Whenever I see a choke in a schematic I just think "big resistor" as I have yet to use one.
As for the new caps being bad I am of the same mind. It seems unlikely. But I wasn't thinking they would make noise themselves as much as possibly fail to filter ripple which I assume would cause audible noise.
We have sound. Not sure yet if I like it but I'll wait until I get the 2x12 cab built before I do any more tweaks.
However I am wondering what the bias should be for KT88s running with 700v plates and 350v screens. Weber and other bias calculators say 35ma for 70% dissipation but AFAIK these all assume that the screen voltage is about the same as the plates. Wouldn't screens runnng much lower change things? Bias is current set at 36ma, no red plating but I'd like to know if I'm runnng things too hot or cold.
When measuring tube current to bias the amp, the screen voltage is already taken into account. Now, if you were trying to bias it by measuring just the bias voltage, then you may well wind up in the wrong place. You've got 700 V x 0.036A, or 25.2 watts of dissipation, which is 60% of the KT88's 42W max. Keep in mind, that if you're measuring the current at the cathode (1R resistor to ground from the cathode), then some of that 36mA is screen dissipation, and your anode current (which is what you're after when adjusting the bias) is going to be a bit less than the 36mA you're seeing.
) with a 270-0-270 400ma HT transformer, if it I run it with a full wave bridge non center tapped I'm not seeing how the tubes could draw more than 400ma. If I was voltage doubling a 270-0 PT with 400ma HT I could understand why that could be a problem. What am I missing?
