• 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.

#26 pre amp

Rod,

Thanks for the advice. BTW, can I use any N-channel MOSFET substitute for your schematic like IRF810/IRF740 (which I have already in my hand)? Also, can I substitute LM4041 with any other voltage reference?

Take care with those FETs! The specified ST part has very low threshold (about 1.7V) versus about 4V for IRFs - you'll have to add extra supply voltage (+4V) to make up for it!

the reference must be adjustable type with 1.2V reference pin - I don't know of any others...
 
Take care with those FETs! The specified ST part has very low threshold (about 1.7V) versus about 4V for IRFs - you'll have to add extra supply voltage (+4V) to make up for it!

the reference must be adjustable type with 1.2V reference pin - I don't know of any others...

Thanks Rod. I'll try to find the same parts or otherwise try my luck with what I have already.
 
Take care with those FETs! The specified ST part has very low threshold (about 1.7V) versus about 4V for IRFs - you'll have to add extra supply voltage (+4V) to make up for it!

the reference must be adjustable type with 1.2V reference pin - I don't know of any others...

Before any change of FET I would be turning the bias down to zero.

IRF in my experience usually need 7 volts bias, 3.5volts times two FETs.
 
Disco,

Those tribute autoformers seem to be favoured, have you managed to compare to any others?

I like the fact that the Silk TVC allows some gain. Both seem to be receiving praises but I haven’t managed to find any comparisons between them on the web.

I hope you’re going to be using mercury wetted relays to complement the quality of the other parts .

Cheers,

Rich
 
Last edited:
Salas, thank you for posting that schematic looks like I have a few Christmas projects.

Coolzero, my B+ power supply of fairly well filtered with chokes and finished off with VR tubes. I’m also planning to replace the series resistor before VR tubes with a ccs, just upgrading the filament supply for now.

I’m starting to put together Rods FET heater, should get my power supply chokes in a few days, Rod thanks for the updated schematic

Marty, good luck with the build. Are those chokes on a steel or ferrite core? The Hammond 60mH 2A is a cheap option, but not common mode. Here is a link to some test results for choke vs cap input.
RR#002 - DC Filament Supply Test

Rich
 
Hi Rich,
Thanks for the link.
I’m not positive on the core material (they are pulls) my guess would be ferrite though.
 

Attachments

  • 100_0828.jpg
    100_0828.jpg
    65.1 KB · Views: 1,032
preregulator filtering for filament supplies

Hi Marty,

They're ferrite cores. If you click on "Lynn’s postings" of that link in my previous post there is more information regarding choke input supplies for filaments along with a discussion. Essentially Lynn Olson suggests 2 chokes (with steel laminations) just after the bridge rectifiers, one in the + rail and the other in the ground (b4 -'ve terminal of the 1st cap). He goes on to suggest a common mode choke, like the ones in a computer psu, to filter high frequency noise.

I'd like to hear your opinion regarding the results you get.

Somewhere in that blog, someone suggests a 3pin regulator wired as a ccs, to simulate the filtering offered by a choke. I also recall seeing this in one of DHT Rob's schematics. It was something like XF Bridge- cap-ccs-cap-ccs-cap ccs- filament. Lynn points out the ccs will drop more voltage than a choke.

Cheers,

Rich
 
26 tubes

Hi all,

I got Blue Arcturus 26 for my pre-amp. Oh-boy those can sing:D..wonderful trebles and mids. So far I tried following tube and below are my observations.

Cunningham Globes- Best overall out of all the tubes I have tested. Extended and controlled bottom and top ends and gorgeous mids. Love those tubes. But little microphonic than other tubes.

Blue Arcturus Globes- Trebles no match to any other 26 tube. But bass is little lean. Least microphonic and detailed sound.

Sylvania ST- Good overall sound. Almost similar to above.

Tung-Sol ST- Nice trebles but highly microphonic. (Not sure it was a problem with TS tube I had)

RCA Globe - Trebles were OK but didn't like the bass. Somewhat bloated and uncontrolled. Very well mate with Arcturus BTW.

Philco ST - Sounds little dry. Less defined than above tubes.

Super Airline ST - Comparatively cheaper tube but sound well worth of trying.


I tried to find CeCo 26 as Kevin suggested but it seems difficult to find. Wonder how it sounds though.:rolleyes:
 
Plate choke or CCS/Gyrator as plate load

Just read a thread that SY explained of using plate choke and CCS as plate loads. Has someone successfully tested both with 26 and compare the sound? Mine I tried both but didn't like the CCS sound much. Not sure whether I have used a well enough CCS. Can some share the experience and explain how both options sound comparatively with 26 tube? Just the sound I am concern about not any other aspect.
 
I converted my 26 preamp, which previously had a IXYS cascode plate load (details in this thread), to use two Hammond 156c back to back, and I think I can say that it's the best option yet. Two of these wired out of phase seem better than just the one - I guess less capacitance = better treble.

It's more up-front than a resistor, better bass, and the main thing is that it's smoother than the IXYS cascode - doesn't have that slight glare. Sound is closer to the IXYS than to the resistor. In the end I thought I was getting some sort of sharpness out of the IXYS, however clear it was (and the bass was terrific). So I went back to a resistor - that was nice and smooth for long term listening, though more distant and less vivid.

So I'd say try the Hammond 156c solution out. I connected the bottoms together, then input to the left wire of the first choke. Connect the right wire of first choke to right wire of second choke, then output from left wire of second choke, so they are out of phase.

andy
 
I converted my 26 preamp, which previously had a IXYS cascode plate load (details in this thread), to use two Hammond 156c back to back, and I think I can say that it's the best option yet. Two of these wired out of phase seem better than just the one - I guess less capacitance = better treble.

I don't have any problem with your preference. To each their own. But I don't think the explanation is accurate. Chokes don't have a phase, per se, so that term is inappropriate. Maybe you could call it "polarity", to reflect which end of the winding is which. Whichever way you connect them in series won't change the frequency response by itself. A particular arrangement of polarity and physical orientation of the chokes, might create small differences, due to some magnetic coupling, which can be additive, or subtractive.

Sheldon
 
Last edited:
Administrator
Joined 2004
Paid Member
I don't have any problem with your preference. To each their own. But I don't think the explanation is accurate. Chokes don't have a phase, per se, so that term is inappropriate. Maybe you could call it "polarity", to reflect which end of the winding is which. Whichever way you connect them in series won't change the frequency response. The resonant frequency will be the same in either case. A particular arrangement of polarity and physical orientation of the chokes, might create small differences, due to some magnetic coupling.

Sheldon

Sheldon makes a rather good point. What you need to do in fact is figure out which end has less overall capacitance to the core and designate that end as the plate connection. The two chokes in series does address the interwinding capacitance issues (which can be very significant) theoretically halving it, but does not address the winding to core capacitance issue at all. Your current orientation is random on this score, whether or not this matters much depends on the construction of the choke. The additional inductance is of course an obvious benefit in terms of LF performance. (one more octave of response before the load impedance starts to become unreasonable.)
 
My simple suggestion:

Three wire power cord with each of the three leads accessible (hot, neutral, ground). From here on out, when I use the term 'ground' I mean this third wire, typically green.

Isolate the core of the choke, maybe lay it on a piece of cardboard.

Connect the two choke leads up to the hot and neutral of the power cord. Note this connection, which lead is connected to hot, which is connected to neutral.

Measure the AC voltage from the choke core to ground. Record as measurement A.

Swap the hot and neutral on the choke (with power off, of course). Repeat AC voltage measurement from core to ground. Record as measurement B.

Compare measurements A and B. The larger number follows the connection where the hot was connected to the higher capacitance lead, and the neutral was connected to the lower capacitance lead. Use the lower capacitance lead as instructed by Kevin above.

This assumes in operation that the choke core is solidly bonded to ground (as it should be).
 
Hi Guys,

When wiring up my Hammond chokes I simply connected them in series and measured the capacitance with a digital multimeter, then swapped the terminals on one choke and re-measured. I then used the option with the lowest capacitance. I must agree with Andy that the dual choke options sounds very good with adequate bass.

An advantage of 2 hammond chokes is that the 300H inductance is significantly greater than most modern line output transformers. Although these measures are at 10mA, I don’t think inductance would double at 5mA. I still haven’t got around to comparing the chokes to my Hashimoto HL-20K.
Hashimoto HL-20K = 100H @10mA
Tango NP-126 = 110H @ 10mA
Silk L941s 10K:600 = 95H @10mA

Cheers,

Rich
 
Hi Guys,

I must agree with Andy that the dual choke options sounds very good with adequate bass.

Cheers,

Rich

I am using EP 200H chokes currently in my 26 pre and sound is far superior to plate resistor or CCS I tried earlier. I always having issues with CCS and couldn't find any improvement with the once I tried so far.

Strangely I felt considerable improvement (better than loading plates with CCS or resistors) with one of the shunt regulator cuircut+plate resistor (22K) combination I tried (tried this because simply wanted to experiment :)). Using one of the Tubecad schematic I have built this shunt regulator. I think it is the 5th schematic in this link Shunt Regulators. Instead of IRF610, I used IRF840 as far as I could remember. This one gave a huge improvement of overall sound with excellent bass control and silky trebles. Do not know what is so special about this combination. Rather, I am not that electronic savvy to figure out what happened exactly. If someone is having an explanation, I am more than happy to know too. Even if I try the same combination now, it is giving the same sound and even consistent with different tubes as well. Hence do not think it was a coincidental scenario. However, the heat dissipation was too much from those resistors (1K / 2.7K) and had to remove it.