Hi
I was looking a really good preamp, and have chosen to build one of the three Jim de Kort designs using the 26 tube at www.vt52.com/diy/myprojects/projects.html (I would be using it in either a 100 dB efficient system or 115 dB horns).
Two of the three of Jim's designs , to overcome the propensity to hum of the 26, use a battery filament supply.
I’m not opposed to batteries, but the other really good option is the TentLabs Directly Heated Tubes Supply www.tentlabs.com/ProductEntities/HeaterSupply/HeaterSupply.html Probably a little more expensive and a little more convenient.
My #1 criteria is sound quality. Has anyone tried both or have a comment about which should sound better?
Thanks!
I was looking a really good preamp, and have chosen to build one of the three Jim de Kort designs using the 26 tube at www.vt52.com/diy/myprojects/projects.html (I would be using it in either a 100 dB efficient system or 115 dB horns).
Two of the three of Jim's designs , to overcome the propensity to hum of the 26, use a battery filament supply.
I’m not opposed to batteries, but the other really good option is the TentLabs Directly Heated Tubes Supply www.tentlabs.com/ProductEntities/HeaterSupply/HeaterSupply.html Probably a little more expensive and a little more convenient.
My #1 criteria is sound quality. Has anyone tried both or have a comment about which should sound better?
Thanks!
My speakers are about 87db efficient and I can run my DHTs in the preamp off bench DC power supplies. I didn't hear any special improvement with batteries. the convenience of DC supplies certainly beats the fuss of recharging batteries. Battery supplies have always been - for me - a kind of "good idea at the time" that I've abandoned through sheer aggravation. There is a rather cool circuit about which automatically switches between two banks of batteries - Gary Pimm or somebody like that. At least that would be a "build it and forget it" option.
I think all this depends on the efficiency of your speakers. Below 90db and you're probably fine with DC supplies. Over 100db and you may well be looking at batteries. I'm looking at common mode chokes in the DC supply as well as voltage regs. Haven't built the supply yet. there's also the RonanReg circuit which is no doubt about on the Net, though John Leverault has run out of boards. Andy
I think all this depends on the efficiency of your speakers. Below 90db and you're probably fine with DC supplies. Over 100db and you may well be looking at batteries. I'm looking at common mode chokes in the DC supply as well as voltage regs. Haven't built the supply yet. there's also the RonanReg circuit which is no doubt about on the Net, though John Leverault has run out of boards. Andy
I'm not qualified to give an educated opinion. I will say this. DHTRob tried one of the Tentlabs for a preamplifier. Remember the Tentlabs were initially made with your typical output tube say a 300B in mind. Rob found they were quite noisy (for a preamp DHT). Guido agreed and declared his intent to improve the tentlabs supply for preamp dht duty, I don't know if he has done that yet. An email to tentlabs can clarify that.
I find the Tentlabs expensive (but that is a fact of life if dealerships are involved) that is the only drawback. Batteries are cumbersome.
You think neither the TentLabs or the RonanReg would be ok over 100 dB?>>
I don't know the answer to that - I've only tried various filament supplies with my 87db speakers, where DC supplies are fine. I know from reading posts that noise becomes a big issue with high efficiency horns, but you'd have to ask people that use them. Andy
I don't know the answer to that - I've only tried various filament supplies with my 87db speakers, where DC supplies are fine. I know from reading posts that noise becomes a big issue with high efficiency horns, but you'd have to ask people that use them. Andy
Why not build your own?
pdf
http://home.quicknet.nl/qn/prive/susan.soeters/TechGraphix/PDF/Electronisch/FilamentSupply.pdf
or if you don't have pdf check out the attached jpg.
This is one Kees Soeters built for his 27 driver tube.
http://home.quicknet.nl/qn/prive/susan.soeters/TechGraphix/PDF/Electronisch/FilamentSupply.pdf
or if you don't have pdf check out the attached jpg.
This is one Kees Soeters built for his 27 driver tube.
Attachments
Bas Horneman said:
You got that right, but in some cases still less cumbersome than trying to put together something that'll work as good. I've never tried Guido's supply modules so I can't speak for them. I ended up with batteries because I'm too lazy to investigate other options - and once chargers and relay control unit is built it's really not that much hustle at least not with my tubes - but they don't draw an amp like the '26 either..
For my 5V/250mA filaments I use 12V batteries and current regulators. Works like a charm.
I think the biggest problem with DC supplies has very little to do with constant current vs constant voltage vs pseudo-whatever, it has to do with the power.
Even if I wanted to use a regulated supply of some kind on DHT filaments I would still power it with a battery. You have a nice amp with a quiet tube full wave rectified choke-input supply (don't you?!), and then you put a diode bridge into a big cap inside it. You'll NEVER get that noise back out of the system, no matter how much you try to filter it.
Gary pimm drew up a system (for the 2v filament tubes) to use a pair of 2v solid lead batteries (made by Hawker, the Cyclon series), where one charges while the other is used, and it switches back and forth periodically. (Every half hour or so, and you don't hear the switch.) With DPDT relays the charger is completely isolated, it doesn't even have to share a ground with your amp. Use a separate small split-bobbin transformer for the timer/charger and you're golden. It's a bit expensive. But if you can't AC heat I really think it's the only option.
You can see it at Gary's old schematics page:
http://www.pacifier.com/~gpimm/schematics.htm
Look at the original '26 choke loaded preamp' and the subsequent PS/charger schematics to see how the relays & charger get wired up.
Even if I wanted to use a regulated supply of some kind on DHT filaments I would still power it with a battery. You have a nice amp with a quiet tube full wave rectified choke-input supply (don't you?!), and then you put a diode bridge into a big cap inside it. You'll NEVER get that noise back out of the system, no matter how much you try to filter it.
Gary pimm drew up a system (for the 2v filament tubes) to use a pair of 2v solid lead batteries (made by Hawker, the Cyclon series), where one charges while the other is used, and it switches back and forth periodically. (Every half hour or so, and you don't hear the switch.) With DPDT relays the charger is completely isolated, it doesn't even have to share a ground with your amp. Use a separate small split-bobbin transformer for the timer/charger and you're golden. It's a bit expensive. But if you can't AC heat I really think it's the only option.
You can see it at Gary's old schematics page:
http://www.pacifier.com/~gpimm/schematics.htm
Look at the original '26 choke loaded preamp' and the subsequent PS/charger schematics to see how the relays & charger get wired up.
Thanks RT, good advice. And thanks for the pointer to Gary Pimm’s site!
Though “Gary Pimm drew up a system (for the 2v filament tubes)” –per the datasheet the 26 requires 1.5 V(how much does 0.5 V matter?)
Gary’s 26 line stages seem (at a quick look m before I head out) to be a tad more straightforward than Kevin’s www.positive-feedback.com/Issue16/kennedy.htm. I wonder which would sound nicer?
Cheers
Though “Gary Pimm drew up a system (for the 2v filament tubes)” –per the datasheet the 26 requires 1.5 V(how much does 0.5 V matter?)
Gary’s 26 line stages seem (at a quick look m before I head out) to be a tad more straightforward than Kevin’s www.positive-feedback.com/Issue16/kennedy.htm. I wonder which would sound nicer?
Cheers
You can see on the schematic he uses resistors to drop the extra voltage. Either a single 0.47, or a pair of 0.22 (one on each leg).
The circuits are very similar, kennedy's uses an output transformer, gary uses constant current sources to load the tube, (if you look at the 26 schem. on the top of the list) then you can take the output from the top of the tube like normal, or from the mu-follower output of the CCS. (the mu-follower output has very low output impedance, but is still limited in output by how much current passes through the tube). Also kennedy uses fixed bias, while gary uses cathode bias.
The biggest differences are in the power supply. Kennedy uses a CRC filter with a series pass voltage regulator after it, Gary uses LCLC filtering with a CCS-fed gas shunt regulator after it. Also Gary uses the battery heaters. I'd say on the supply front Gary's design is far superior. But, more complicated and more expensive. There will be less parts in the signal path, which is always good.
As for the circuit itself, it depends a lot on what you're driving. Does the amp have 10K input impedance (common for solid state) or 50-100K (common for tube amps). Do you need the full voltage gain of the 26, or not?
Building an absolute-best (ccs-fed shunt reg supply w/ ccs loads and battery heaters, imo) 26 preamp may not be the ideal first project, though. You could start with a simpler rCLC supply, a couple gas tubes, and some easy kit CCS's like the bottlehead C4S. Then do a DC current heater supply with some LT1085's. You could always upgrade to the tent labs supplies later if you wanted to, or batteries.
The circuits are very similar, kennedy's uses an output transformer, gary uses constant current sources to load the tube, (if you look at the 26 schem. on the top of the list) then you can take the output from the top of the tube like normal, or from the mu-follower output of the CCS. (the mu-follower output has very low output impedance, but is still limited in output by how much current passes through the tube). Also kennedy uses fixed bias, while gary uses cathode bias.
The biggest differences are in the power supply. Kennedy uses a CRC filter with a series pass voltage regulator after it, Gary uses LCLC filtering with a CCS-fed gas shunt regulator after it. Also Gary uses the battery heaters. I'd say on the supply front Gary's design is far superior. But, more complicated and more expensive. There will be less parts in the signal path, which is always good.
As for the circuit itself, it depends a lot on what you're driving. Does the amp have 10K input impedance (common for solid state) or 50-100K (common for tube amps). Do you need the full voltage gain of the 26, or not?
Building an absolute-best (ccs-fed shunt reg supply w/ ccs loads and battery heaters, imo) 26 preamp may not be the ideal first project, though. You could start with a simpler rCLC supply, a couple gas tubes, and some easy kit CCS's like the bottlehead C4S. Then do a DC current heater supply with some LT1085's. You could always upgrade to the tent labs supplies later if you wanted to, or batteries.
Bas Horneman said:
Hi,
The low noise versions are available, they are 129 euro a pair (including VAT) and are 28dB more quiet than the initially launched versions for output tubes.
Bear in mind that not only the low noise, but also the high output impedance makes that these units sound so transparent and clean.
About the pricing: I'd wish to make them cheaper but believe me that I make less money compared with my employed days at Philips......
best
Rescue Toaster said:
Hi,
We (and others) found that the impedance across the filament affects the sound. For background, see:
http://www.tentlabs.com/Info/Articles/Heatingmethods.pdf
could you eleborate on what you mean with "it has to do with power"
best
Yes, you're right, if it's DC it should be a high impedance source.
I just think that with a diode bridge going into a capacitor, you're never going to get that noise back out of the system. You can use shottkys and snub it and so forth, maybe with choke input... but I'd need to see some real frequency analyzer output before I'd put a diode bridge into the circuit.
Or just run the regulator off batteries. I'm less concerned that your regulator is bad (I'm sure it's damn good) than I am with what I see as a nearly fundamental problem with ac-powered dc heating.
Sure you can take steps, get the diode switching/cap charging noise low, but how low?
I just think that with a diode bridge going into a capacitor, you're never going to get that noise back out of the system. You can use shottkys and snub it and so forth, maybe with choke input... but I'd need to see some real frequency analyzer output before I'd put a diode bridge into the circuit.
Or just run the regulator off batteries. I'm less concerned that your regulator is bad (I'm sure it's damn good) than I am with what I see as a nearly fundamental problem with ac-powered dc heating.
Sure you can take steps, get the diode switching/cap charging noise low, but how low?
Rescue Toaster said:
Hi
Yes, batteries are cleaner than rectified AC, though not all batteries perform equal......
On noise: My low noise version has about 8nASqrrtHz output noise. Depending on the filament resistance, that translates into a noise voltage. It is not that difficult to obtain similar noise specs on a voltage regulator.
best
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.