• 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.

all dht se PX4

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
On high efficiency speakers (95dB probably) no hiss or humm. Notice the distance between the power tranny and the #26...

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.


Voices and piano to die for [uhrr... in a positive way that is],
despite the bad room acoustics in the shack.
 
Hi Andy,

You'd suggest to use the negative filament as the cathode point and loose the resistors instead?

Well, might try that ;)
The internal resistance of a real lead cell is quite low,
the change will have less impact than with a linear reg I presume.

That's a C-core 1:1. BW > 35KHz @ 10mA if I recall correctly.
What's the improvement of the 4P1L?
 
Last edited:
Did some optimizing (by ears). Will do the calculations next to see if improvements are possible. There always are, not?

An externally hosted image should be here but it was not working when we last tested it.


Some overall feedback is applied to help the damping factor, it reduced some of the outputs' nastiness too. From a clean sinus wave I could not hear a difference in output level < 15mV. 25mV level difference was hard to notice, it depended on frequency.

An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


I used cheap iron, better grade will be tested as well as a shunt regulator for the driver stage. Sounds already pretty good from what I'm hearing right now.
 
Was thinking today about the double voltage amplifier setup and figured that the KC1 could do the work alone, leaving the 26 for buffering service. Tinkering with the load line I settled for 57K, producing circa 12Vrms at the expense of distortion. That we don't want and besides the 34Vpp was not enough to drive the PX4 to full output. Are there options unexplored? Well, yes (there always are, not?) an interstage under the cathode of the 26.

Have you tried this already 26-gang?

It must be an impedance thing, because the AE -which is a fine 1:1 interstage- didn't keep up with the former reproduction quality. I had another shot with an 1:2 interstage and -bingo- that 5K to 20K did the trick. In went a Tamura B-6004.

An externally hosted image should be here but it was not working when we last tested it.
 
Just an observation : you should remove the cathode connection from the resistor centre tap , remove the resistors (which do nothing apart from run down the batteries) and tie the cathode connection to one side of the filament

I found 3C24 was a super , single stage driver for Osram PX4 . It's a pain in the neck to light up and needs it's own (higher) HT supply but IMO is worth it and sounds superb with PX4/2A3 etc :)

316A
 
Beautiful valve. Guess that amp has a strong industrial design to it.
The reason for my low supply voltage is in the 30's era bulbs I'm using. Those oxide coated battery tubes are very linear small signal tubes. KC1 has high mu at the expense of high Ri, so possibilities are limited -hence the buffer stage.

An externally hosted image should be here but it was not working when we last tested it.


The setup I'm working on now:

An externally hosted image should be here but it was not working when we last tested it.


Yes, there should be a decoupling cap at the anode of the 26 ;)
 
Last edited:
Something fishy crept in. On high levels there was a sharp edge to the midrage, formerly not heard. The last change was the Tamura B-6004 interstage so I opened the manual: 30Hz to 15KHz within 4dB, output power level 25dBm. I understood the power level having to do with the bandwith, but it seems to do with the onset of distortion. Lacking a professional power meter I exchanged the interstage for other models and finally for a coil and coupling capacitor. The sharp edge was gone (those C-cores are spaced for 50mA) but every coil or interstage brought their own characteristic "sound". Although I could not stand the distortion of the B-6004, I missed that lush low end. So much that I reinstalled and only used its primary. Coupled with an oil cap I have the impression I'm near finishing. The reproduction is at a level I wish everyone could enjoy, a feast to the ears. :note::note::note:

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.
 
Just added the Coleman Filament Supply to the PX4 so the only unregulated power source now is the 228 volt HT. If you're not familiar with DC heating DHTs with Rods' fruit of labor, then you're in for a shock installing one. One? Buy 'm at a dozen because they're a steal!

An externally hosted image should be here but it was not working when we last tested it.


What may you expect, crossing over from the dark side?
From the first seconds I missed that 'romantic' feel about the music, leaving room for imagination. Now small details are on top of the bigger sounds, making the music easy to absorb, very precisely. Thinking about a parallel experience, changing the bed sheets comes to mind (every odd year) :)
Chapeau sir, wel done!

Disco's off to the woods now, drawing crosses on ancient oaks.
 
The Triode Guild® was a great inspirator for poetry ;)

Well, to get rid of that one pesky electrolytic cap I'd burn an extra 20 watt, if my calculation is right:

An externally hosted image should be here but it was not working when we last tested it.


Rb= 24V/1,040A= 23 ohm
P= 29 watt

In the actual setup the filament supply has a rendement of 4.0W over 8.5W, that's a figure I can live with :)
 
chop chop, a new case, a new built.

The PS was built round one salvaged Gardner transformer, offering better regulation and higher supply voltage. Feeding the PX4 filament took a few extra turns of 1mm wire. Unfortunately the PCB of the HV reg needs to be revised for extending with kelvin connections. So it feeds a 30cm remote load via shielded wire, not an ideal situation but compromises had to be made. I found that tinning the PCB traces had a positive influence on temperature dependency. It sets and holds (at 60mA) as a charm.

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.


The idea was to squeeze all in to a small box, to be acceptable for permanent use in the living room. It measures 31x21cm and is made from solid, fine sanded wood (Finnish birch -Betula pubescens- and Indies Rosewood -Sonokling-) adding on to the delightfull aroma of ancient tubes. The bottom is open for ventilation, the lower front is made of 3mm copper to cool the semi's of the LV regulator. The meter is an indication for HV because the mains has its ups and downs. I could not get rid of that very ugly elektrolyte cap (a 350V Cerafine 2x 220uF, also decoupling the PX4 cathode) because the OPT is taking up most space underneath. The lead cells are to come under the case, in a seperate plinth.

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.


When under development I wandered between two thoughts: a buffered voltage amplificator versus a 'normal' two stage anode follower. For practical reasons (swing at the PX4) I opted for the last, leaving the cathode follower to be examined perhaps on a later time -it certainly offers its own charms. The KC1 triode multiplies the theoretical 0.7Vrms output level of a CD-player eleven times while the #26 in this setup amplifies three times. Full throttle the PX4 receives circa 30Vrms, a little more than the targetted 60Vpp.

An externally hosted image should be here but it was not working when we last tested it.

The first coupling cap reads 0.1Uf, it should be 1.0uF ;)


Reading here and there about problems asociated with the use of type 26 (rattle and humm, microphonic or ringing noise) I experience none at all on the twenty nine 26 tubes I own, from well used thirties globes to unused fifties ST. Without restraint I hold the lead cell accountable for this feat as my circuit is probably like yours.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.