• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

transformer based phase splitter

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I have this UTC O-6 which has primary 15K into 95K CT secondary. Is this an appropriate trans for phase splitting. If so what kind of tubes would be best suited for this ? I have mostly pentodes and beam pentodes, el84 to kt120 in terms of power output. I also have output trannys to match. Thank you for your suggestions. cheers, 808
 
I just tested quite similar high impedance audio transformers and noticed that good frequency response require some amount of testing, impedance matching and testing again....
I could not get satisfactory results. Especially transient response was a mess with a lot of ringing.
 
This is how I did it. It sounded pretty good, but lots of ringing on square waves even with gNFB. By the time I increased the cap in the gNFB network to 7500pf the ringing had stopped, along with high frequency response. I gave up and ended up using a split load phase splitter instead.
 

Attachments

  • 25790897_10155785305716006_1226196334886724409_o.jpg
    25790897_10155785305716006_1226196334886724409_o.jpg
    51.8 KB · Views: 478
This is how I did it. It sounded pretty good, but lots of ringing on square waves even with gNFB. By the time I increased the cap in the gNFB network to 7500pf the ringing had stopped, along with high frequency response. I gave up and ended up using a split load phase splitter instead.

Why? Make it simpler and put the bias voltage on the secondary CT. Remove the coupling caps to the power tubes. Remove both 100K resistors to power tube grids.
 
I used Hammond 124B, and I biased it this way because I used individual bias controls for each tube since the OPT was a power toroidal. Using a single voltage for both tubes results in too much imbalance.
Yes, the IT secondary would need to be a split winding terminal format to allow some independent biasing adjustment to be incorporated.
 
I have this UTC O-6 which has primary 15K into 95K CT secondary. Is this an appropriate trans for phase splitting. If so what kind of tubes would be best suited for this ? I have mostly pentodes and beam pentodes, el84 to kt120 in terms of power output. I also have output trannys to match. Thank you for your suggestions. cheers, 808

Since the IT has low input impedance and steps up, I'd use a medium mu triode such as 6SN7 or 6DJ8 etc. Or how bout trioded EL84s? What mu do they have as triodes?

I've been using EDCORs XSM series as IT and am very happy with them. I do not use feedback around them tho, have never felt the need for GNFB. But thats just me.
 
I used Hammond 124B, and I biased it this way because I used individual bias controls for each tube since the OPT was a power toroidal. Using a single voltage for both tubes results in too much imbalance.

Regarding ringing you cant tame...I ignore the square wave tests when using ITs. It sounds so good who cares...

U could modify the IT by desoldering the center tap and getting the secondaries separated. I've done that on EDCOR transformers. Just do it before too many cups of coffee, and fasten the wires to new pins before measuring the results, the wires are so fragile and will break if bent more than a few times. But once they are scured to a pin they are ok to work with.

Then u could do like this:
 

Attachments

  • IT_mod.PNG
    IT_mod.PNG
    69.9 KB · Views: 429
I've never been satisfied with interstage transformers other than the rare exceptions where I've been fortunate enough to work with very expen$ive iron... They tend to ring even without being wrapped in a feedback loop, and can be a bear to tune at times. I would prefer even a *shudder* floating paraphase splitter to an interstage except in very rare instances, or at line level. Since I no longer have an oscilloscope to tune them, I don't fool with them anymore.


That having been said, they are a period correct way to go for AB2 DHT types of builds, and if you can tame them, they work well.
 
I think a small trafo like the O6 might be better at the input. I would try it into a diff pair. Need to watch input C of the next section though. Might make a tough load for the preamp.

Agreed. I think an ECC99 in the diff. amp will handle just about anything downstream. IIRC, Lynn Olsen did this very thing.

To avoid complications, employ good O/P "iron" and stick to short loop NFB, within each leg. All "iron" will be outside of any NFB loops.
 
I think a small trafo like the O6 might be better at the input. I would try it into a diff pair. Need to watch input C of the next section though. Might make a tough load for the preamp.

So, by having the input going to pin 1, p2 to ? . on the secondary the p3 to one tube and p4 to ? then p5 to the other tube. Or both outputs to 2 different sections of a dual triode. You don't happen to know of a schematic of an amp where this type of IT is used as an input splitter.
 
Agreed. I think an ECC99 in the diff. amp will handle just about anything downstream. IIRC, Lynn Olsen did this very thing.

No.
Lynn used a line level input transformer; all next stages are transformer coupled class A push pull.
Phase splitter transformers at "interstage" level are real nightmares to get right; I experimented a lot.
Measurement wise, line level input transformers using small high permeability cores and without any DC currents through windings are much much better.
Of course some will prefer the "sound" of the inferior interstage phase splitter transformer :rolleyes:
 
As previously stated, Lynn Olson did work along these lines. Look here. Lynn too warns of the very difficult load this sort of circuitry presents to the driving electronics. Use a big, honking, MOSFET source follower running at about 30 mA. of drain current ahead of the phase splitting trafo. The FET will effortlessly charge and discharge the nasty net capacitance, while exhibiting an O/P impedance well under 100 Ω. Remember that trafos are inherently low impedance devices and the low impedance driver will squeeze the best possible linearity out of the "iron".
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.