Because I was bit lazy and in a hurry to complete my pre, I used something similar to below. It turned out to be a complete success with zero hum.
In fact it is a cheaper option too. But no fun like build one your own, if you have time.
. I used the 5.5V reduced to 3 V and filament in series. Bit of a cross talk and channel imbalance specially when you are not using matching tubes, but I was quite happy with the outcome. You could use 2 of these for each channel to avoid such issues by referencing the voltage to 1.2V.China Switching Power Supplies. Mianyang Manufacturer, Importer, Exporter
Probably you haven't read properly what I have put
. I haven't said Blue globe is "the tube"; rather I mentioned that, it has a superior highs than others out of the once I heard. In fact my best overall choice was Cunningham globes as I mentioned. Anyway, sound of a particular tube is always subjective and no point of arguing about that. Whoever wish to experiment, should listen to different tubes and choose whatever the best for their taste. But there is no harm I believe of sharing the experience with others.Hi Coolzero,
There may have been a misinterpretation in that point, since I didn’t add the obligatory emoticon (Don’t listen to Coolzero, the blue globe 126 are not that special).
A few days after your post on 26 comparisons there were a few Arcturus blue 26’s appearing on ebay that ended up rocketing in price, I still need 1 more to replace a bad one I have.
I’ve also tested the valves you have reported on and a few others. I would agree with most of your findings, which have been very useful. As for posting which ones I reckon are best overall, I’ll do so after I have a sufficient stash. Shhh about the Cunninghams we don’t want to price to reach that of the Arcturus.
Cheers,
Rich
There may have been a misinterpretation in that point, since I didn’t add the obligatory emoticon (Don’t listen to Coolzero, the blue globe 126 are not that special
). A few days after your post on 26 comparisons there were a few Arcturus blue 26’s appearing on ebay that ended up rocketing in price, I still need 1 more to replace a bad one I have.
I’ve also tested the valves you have reported on and a few others. I would agree with most of your findings, which have been very useful. As for posting which ones I reckon are best overall, I’ll do so after I have a sufficient stash
. Shhh about the Cunninghams we don’t want to price to reach that of the Arcturus. Cheers,
Rich
Hi Coolzero,
There may have been a misinterpretation in that point, since I didn’t add the obligatory emoticon (Don’t listen to Coolzero, the blue globe 126 are not that special
A few days after your post on 26 comparisons there were a few Arcturus blue 26’s appearing on ebay that ended up rocketing in price, I still need 1 more to replace a bad one I have.
I’ve also tested the valves you have reported on and a few others. I would agree with most of your findings, which have been very useful. As for posting which ones I reckon are best overall, I’ll do so after I have a sufficient stash
Cheers,
Rich
..Rich, until you told, I haven't noticed much of what had happened on ebay. Luckily I got hold of a single blue globe in good condition, few days before my post. Now I know why I couldn't got hold of any other blue globes afterwards.
Never could find sound impressions of this say in comparison to the CCS solutions.
Hi everyone:
I am glad I have found my way to this thread about the 26 preamp here. I feel for most of the audiophiles here on this site of the world, it's a kind of a LOVE and HATE thing. As soon as one encounters the filament heater issue, one would bravely dig out a voltage regulated supply, handle it over to me and close the doors... Not that it is bad, but as one would say, no pain no gain.
Therefore, I want to say "thank you" to everyone who has contributed to this thread. I have been with the idea for 3 years to build it and now that I have almost all of the parts, think, should give it a go if it was not for two reasons.
- I have been running around with the idea of "1:4 step up", "26", "4:1 step down", but think lately that I will listening to the signature of transformers rather than the true DHT triode 26. A plate choke or just simply the "4:1 step down" is back in the picture. Question : what is the best option to hear the true "26" that people are talking about.
- Then comes the heater issue : Rod, thank you for the circuits you are sharing with us. Can you tell me which is the final circuit (FET ?) you suggest for the 26 preamp ?
Your gyrator/CCS is inspiring me to resolve another problem I am dealing with currently : I have just built a Parallel Push Pull 2a3 power amp monoblocks (4 tubes 2a3), cathode biased, feeding with pure AC !!! I am relying on the matching of the 4 tubes, a bit silly, now with audible hum on my low efficient speakers 87 dB. Any help to feed those 4 heaters with your gyrator/CCS ?
Thanks in advance.
NG
I am glad I have found my way to this thread about the 26 preamp here. I feel for most of the audiophiles here on this site of the world, it's a kind of a LOVE and HATE thing. As soon as one encounters the filament heater issue, one would bravely dig out a voltage regulated supply, handle it over to me and close the doors... Not that it is bad, but as one would say, no pain no gain.
Therefore, I want to say "thank you" to everyone who has contributed to this thread. I have been with the idea for 3 years to build it and now that I have almost all of the parts, think, should give it a go if it was not for two reasons.
- I have been running around with the idea of "1:4 step up", "26", "4:1 step down", but think lately that I will listening to the signature of transformers rather than the true DHT triode 26. A plate choke or just simply the "4:1 step down" is back in the picture. Question : what is the best option to hear the true "26" that people are talking about.
- Then comes the heater issue : Rod, thank you for the circuits you are sharing with us. Can you tell me which is the final circuit (FET ?) you suggest for the 26 preamp ?
Your gyrator/CCS is inspiring me to resolve another problem I am dealing with currently : I have just built a Parallel Push Pull 2a3 power amp monoblocks (4 tubes 2a3), cathode biased, feeding with pure AC !!! I am relying on the matching of the 4 tubes, a bit silly, now with audible hum on my low efficient speakers 87 dB. Any help to feed those 4 heaters with your gyrator/CCS ?
Thanks in advance.
NG
The gyrator/CCS circuit was designed to improve the sound of 300B amps. With 2A3, you should hear much less hum AND much better sound. If you use the Sovtek version, the difference could be dramatic (it was with Sovtek 300B).
The design is straightforward, but you have to think about the heat dissipation. Each side of PP needs 5A, and you should aim to have 2V across each FET. Use four in total (2 CCS, 2 gyrator) So you're burning 40 to 50W total in 4 FETs.
STP55NF06L can handle 95W each, and has thermal resitance 1.6K/W. The heatsink should be 0.5 to 0.7K/W to prevent anyone burning their fingers on it! That's quite a lot of heatsink, which must be in free air to work right.
STP55NF06L is a logic level FET, so 2 to 2.5V across it will be fine.
Hi NG,
If you have the output transformer go for it, otherwise just build the plate choke version. Time spent not listening to a DHT 26 preamp is time wasted, in this life.
I’ll build Rod’s filament when I get time, but the LM350 ccs is fine to start with. Use choke input if you can or small cap, resistor then big cap and ccs.
The only major concern is finding quite tubes, although all 26’s are created equal, some are more equal than others and some are more microphonic than others.
Cheers,
Rich
PS, I don't recall anyone raving about Silvertone 26 globes, must be better than Arcturus!
http://cgi.ebay.com.au/Qty-3-Globe-...ptZVintage_Electronics_R2?hash=item5ad7e1eb6c
If you have the output transformer go for it, otherwise just build the plate choke version. Time spent not listening to a DHT 26 preamp is time wasted, in this life.
I’ll build Rod’s filament when I get time, but the LM350 ccs is fine to start with. Use choke input if you can or small cap, resistor then big cap and ccs.
The only major concern is finding quite tubes, although all 26’s are created equal, some are more equal than others and some are more microphonic than others.
Cheers,
Rich
PS, I don't recall anyone raving about Silvertone 26 globes, must be better than Arcturus!
http://cgi.ebay.com.au/Qty-3-Globe-...ptZVintage_Electronics_R2?hash=item5ad7e1eb6c
Last edited:
C2 not critical. reduce it all the way down to 470n if that's what you have at hand. Experiment with increased R3 if you do that. The C2/R3 filter is there to reduce hum level at the filament even further. The sound improves even when hum is inaudible, though.
R2 is burning about 1,5W for the 26, but you don't want it getting hot at all. I like the 20W TO-220 resistors here (see Farnell), attached to to chassis or heatsink. The others are low power, so any nonmagnetic will do - I use 1206 chip resistors so that the circuit can be built small - this reduces the chances of EM pickup of any kind, which is important in any audio circuit.
R2 is burning about 1,5W for the 26, but you don't want it getting hot at all. I like the 20W TO-220 resistors here (see Farnell), attached to to chassis or heatsink. The others are low power, so any nonmagnetic will do - I use 1206 chip resistors so that the circuit can be built small - this reduces the chances of EM pickup of any kind, which is important in any audio circuit.
He did not relay any news to me either since his last posts. Maybe he did not make the comparisons he was about to, yet.
Hi Rod:
Thanks for your advice. Just a little bit more about my PPP 2A3 : they are indeed Sovtek, matched quads for each monoblocks. I have connected the four 2A3's heaters in parallel on each monoblock and feeding them with 2.5 V tap directly from my PTX. Assuming that each heater takes up 2.5 A, you mean to say that the 4 FET (2 FET for gyrator and 2 FET for CCS) will be enough to handle the 10 A current, is that right ? Any other modification in your scheme ?
Second question, since I am using cathode bias, should I connect the cathode bias resistor to the "filament supply 0 V" of your circuit ?
Hi Richard:
Thanks for your advice too. I share your idea that I should start listening ASAP instead of continuously looking for the best circuit.
By the way, I will be getting a Silvertone ST triode 26 soon. Unfortunately, it is only a single one, but the day I have hear it, I will let you know how it sounds in comparison with the globe RCA/Cunninghams I have collected.
Regards,
NG
Thanks for your advice. Just a little bit more about my PPP 2A3 : they are indeed Sovtek, matched quads for each monoblocks. I have connected the four 2A3's heaters in parallel on each monoblock and feeding them with 2.5 V tap directly from my PTX. Assuming that each heater takes up 2.5 A, you mean to say that the 4 FET (2 FET for gyrator and 2 FET for CCS) will be enough to handle the 10 A current, is that right ? Any other modification in your scheme ?
Second question, since I am using cathode bias, should I connect the cathode bias resistor to the "filament supply 0 V" of your circuit ?
Hi Richard:
Thanks for your advice too. I share your idea that I should start listening ASAP instead of continuously looking for the best circuit.
By the way, I will be getting a Silvertone ST triode 26 soon. Unfortunately, it is only a single one, but the day I have hear it, I will let you know how it sounds in comparison with the globe RCA/Cunninghams I have collected.
Regards,
NG
Trying to design the PPP DHT heat is harder than it looked!
The negative side of the filament should be connected to the cathode resistor. This dictates a P device providing current source. Looking at the available FETs that will operate at 2V or so Vds (factoring in Rds(ON) at 5A) the capacitance is an absurd 500pF or more at the working voltage, which is going to let all the HF noise straight through!
A PNP transistor will give much better performance, but you must choose carefully. The Toshiba 2SA1887 will deliver 5A at 1V Vce, and 40mA base current.
Attempting the gyrator at 10A would be tough. Probably best to find a drum-cored choke that can take the current and use that instead. The diagram tells the tale.
The negative side of the filament should be connected to the cathode resistor. This dictates a P device providing current source. Looking at the available FETs that will operate at 2V or so Vds (factoring in Rds(ON) at 5A) the capacitance is an absurd 500pF or more at the working voltage, which is going to let all the HF noise straight through!
A PNP transistor will give much better performance, but you must choose carefully. The Toshiba 2SA1887 will deliver 5A at 1V Vce, and 40mA base current.
Attempting the gyrator at 10A would be tough. Probably best to find a drum-cored choke that can take the current and use that instead. The diagram tells the tale.
Attachments
Last edited:
Boy, after struggling with this same issue (and eventually resorting to standard LT1086 CCS heating), I really have my doubts that any BJT on the planet can accomplish this.
I recall my power electronics pseudo-professor (he was actually a design engineer in the real world) saying he always designs his BJT circuits assuming a beta of 10. Period, no exceptions. After much frustration, I am starting to agree with this formula. Especially when low capacitance is an important performance parameter.
That would be one heckuva transistor to accomplish this at even poor capacitance performance.
Hi Rod:
Thanks aaaahh lot ! I hope very soon I can finally start listening to some music after 2 years of building my PPP amp. The 26 preamp later on should be the "crown" !
Forgive me to ask this silly question, but I don't understand why there is only 1 drum cored choke L1 ? What should be the spec of the choke ?
Once again, thanks for your help.
Regards,
NG
Thanks aaaahh lot ! I hope very soon I can finally start listening to some music after 2 years of building my PPP amp. The 26 preamp later on should be the "crown" !
Forgive me to ask this silly question, but I don't understand why there is only 1 drum cored choke L1 ? What should be the spec of the choke ?
Once again, thanks for your help.
Regards,
NG