AC does sound better than those 3-pin regulators but there exists something totally different. AC still sounds bad!
I'm not an advertiser here, nor I am receiving any credits from Rod.
There is no need for you to re-invent the wheel. I am a coil winder and have experimented with the most insane solutions, go back several pages and just read what I've done here...
I have listened to anything you could imagine of. Rod's circuit gave me something else but not perfect. With this simple mod it surpassed all my expectations.
Do not lose your time but only if you would like to experiment different flavors. If you want to go to the end, there is no reason to lose your time.
I've already done it for you...
No placebo effect here!
Everybody is free to do whatever he wants.
We are all well skilled DIYers, I also agree on this!
I just wanted to pass my 2 cents here. Nobody is perfect, we're all here to cooperate towards illusionary perfection. But please let's keep this thread up to the high level it deserves.
Greetings from Athens!
Hi All
Firstly I’d like to praise Rod’s filament heater supplies. I’m using 300B version in a couple of SE amps, and there are unbelievably good. I went from an unregulated dc supply, then tried some dedicated CCS regs from DIY HiFi Supply- much better than the dc supply- to Rod’s circuit. There’s nothing subtle about the improvement they bring, and, if you can accommodate them you should! (You should also try his shunt cascode driver topology, which, if anything, is even more amazing- but I digress.)
Secondly, I’m quite open to the idea that changing the MOX resistor in the regulator for a carbon type. Generic carbon film resistors sound much better generic metal films IMO (unless high tolerance is a circuit requirement), despite many theoretical reasons for the reverse to be true.
But I’m a bit confused about the graphite rod business. I’ve done a bit of research, and it seems we’re dealing with two different materials. Graphite rods (for electrolysis etc) seem to be made solely from graphite, which is compressed into a rod-
Graphite & Carbon Technical Data
But these materials have resistivities in the range ~7 (medium grade) to ~ 15 (fine grade) ohm.mm2.m-1 which means than even a narrow, 3 mm diameter rod would have to be ~20- 40 cm long to make a 0.82 ohm resistor! Which suggests the material Peter used was in fact a pencil lead- made form a baked mixture of graphite and clay. I.e. a graphite /ceramic mixture- as used in carbon composition resistors- which he hates the sound of!
Whatever. If we want to try suitable resistors made up of either carbon composition, or carbon film ( I think these use pure carbon rather than a mixture, but I may be wrong), choice is a bit limited. High wattage, low ohmic types aren’t made, and even values below 10 R are rare. I think I’m going to try carbon films- 4 x 3R3 1W types in parallel-
MULTICOMP|MCF 1W 3R3|RESISTOR, 1W 5% 3R3 | Farnell United Kingdom
For composition types we could try 4 of
OHMITE|OA33GKE|RESISTOR, CARBON, 3.3OHM, 1W, | Farnell United Kingdom
but these are pricey (especially in the UK)
Comments? Please be gentle if I’ve missed something here
Paul
Firstly I’d like to praise Rod’s filament heater supplies. I’m using 300B version in a couple of SE amps, and there are unbelievably good. I went from an unregulated dc supply, then tried some dedicated CCS regs from DIY HiFi Supply- much better than the dc supply- to Rod’s circuit. There’s nothing subtle about the improvement they bring, and, if you can accommodate them you should! (You should also try his shunt cascode driver topology, which, if anything, is even more amazing- but I digress.)
Secondly, I’m quite open to the idea that changing the MOX resistor in the regulator for a carbon type. Generic carbon film resistors sound much better generic metal films IMO (unless high tolerance is a circuit requirement), despite many theoretical reasons for the reverse to be true.
But I’m a bit confused about the graphite rod business. I’ve done a bit of research, and it seems we’re dealing with two different materials. Graphite rods (for electrolysis etc) seem to be made solely from graphite, which is compressed into a rod-
Graphite & Carbon Technical Data
But these materials have resistivities in the range ~7 (medium grade) to ~ 15 (fine grade) ohm.mm2.m-1 which means than even a narrow, 3 mm diameter rod would have to be ~20- 40 cm long to make a 0.82 ohm resistor! Which suggests the material Peter used was in fact a pencil lead- made form a baked mixture of graphite and clay. I.e. a graphite /ceramic mixture- as used in carbon composition resistors- which he hates the sound of!
Whatever. If we want to try suitable resistors made up of either carbon composition, or carbon film ( I think these use pure carbon rather than a mixture, but I may be wrong), choice is a bit limited. High wattage, low ohmic types aren’t made, and even values below 10 R are rare. I think I’m going to try carbon films- 4 x 3R3 1W types in parallel-
MULTICOMP|MCF 1W 3R3|RESISTOR, 1W 5% 3R3 | Farnell United Kingdom
For composition types we could try 4 of
OHMITE|OA33GKE|RESISTOR, CARBON, 3.3OHM, 1W, | Farnell United Kingdom
but these are pricey (especially in the UK)
Comments? Please be gentle if I’ve missed something here
Paul
Glad you're enjoying the 300Bs, Paul, and thank you for the recommendation.
Peter's experiments have set me down this route, too.
The R1 current sense resistor in the Filament Regulator Kits is a Wirewound - 0,82 for 300B. This value is not critical, since the trimmer is there to accommodate differences.
Please note that the Wirewound I supply is FLAMEPROOF, that's to say that it won't set fire to anything even if you short the whole regulator out or something similar. The action of the regulator prevents short circuits, so this is a vanishingly unlikely event, but please be warned.
about 30 minutes ago I set up a 300B regulator with 2x 1,5-ohm carbon film 1W resistors. RS 7078558. They are notable straight away for running very cool, even at 1,3A (about 0,6W each). I'll run them steady for a while, then look at the noise, and maybe try them in my 300B-SE amp.
If you want to try Carbon comp, then the low-cost way to go is Farnell 0,5W 6,8-ohm - parallel 8 pcs. These are only about 10p (~6 cents) a copy.
I think that the challenge with Carbon/graphite rods will be the lead-attach. We need a good grip, but without shattering the rod. Preforming on an undersized rod might be one route.
Peter's experiments have set me down this route, too.
The R1 current sense resistor in the Filament Regulator Kits is a Wirewound - 0,82 for 300B. This value is not critical, since the trimmer is there to accommodate differences.
Please note that the Wirewound I supply is FLAMEPROOF, that's to say that it won't set fire to anything even if you short the whole regulator out or something similar. The action of the regulator prevents short circuits, so this is a vanishingly unlikely event, but please be warned.
about 30 minutes ago I set up a 300B regulator with 2x 1,5-ohm carbon film 1W resistors. RS 7078558. They are notable straight away for running very cool, even at 1,3A (about 0,6W each). I'll run them steady for a while, then look at the noise, and maybe try them in my 300B-SE amp.
If you want to try Carbon comp, then the low-cost way to go is Farnell 0,5W 6,8-ohm - parallel 8 pcs. These are only about 10p (~6 cents) a copy.
I think that the challenge with Carbon/graphite rods will be the lead-attach. We need a good grip, but without shattering the rod. Preforming on an undersized rod might be one route.
Hi Rod, we just now are getting settled in our new house and so I have time to start back up on the GM70 amp project. I am thinking of putting some of the PS (including the GM70 filliment PS) in a separate chassis. Is there going to be an issue with running a foot or so of umbilical chord to the main amp?
You're doing the right thing, for sure!
Keep the transformer/caps/rectifier in the PSU chassis, and run the raw dc to the signal chassis. The regulator should be within about 150-250mm (6-10") of the tube socket. Put 1000uF/35V on the PCB supply input terminals, and let us know how it sounds!
Keep the transformer/caps/rectifier in the PSU chassis, and run the raw dc to the signal chassis. The regulator should be within about 150-250mm (6-10") of the tube socket. Put 1000uF/35V on the PCB supply input terminals, and let us know how it sounds!
Thanks Rod. As it is designed now, the main B+ is on the main chassis, but I still need room for the GM70 filly supply, 6.3v for the 12GH7, 5v for the rec tubes, and 350vB+ for the 12GH7. When I looked at the main chassis there just is not that much room. I'll let you know how it turns out.
A little add-on here...
If it is unavoidable to use capacitors in both two chassis, try not to parallel them. Paralleling power supply capacitors either on filament or on HV supply does something really bad on sound. Do not place caps on both chassis. I would prefer the diodes on one chassis and the caps on signal chassis. Best solution is of course to use a pi filter C-L-C with the first CL section on PS chassis and C with the regulator on signal chassis.
Just my two cents...
One more surprised victim here
Yesterday I finished the DHT regulators. Last stop was Rod's solution. I have tried crazy and out of this world ideas there, even tripple chokes, two differential and one monstrous common mode right before the filament.
Rod's circuit gave something I was missing till now. When switching back to common mode choke (22H of inductance!) I felt I missed a great part of immediacy and dynamism of music. I will try to replace 1-2 parts with better sounding ones, I think it deserves so because the result is fantastic.
I will be back with more news during the days ahead.
Immediacy and dynamism of music, you got that right mate!
I owe this post to Rod as I promised him a review of his regulators sometime last year after implementation of it in my various customized DHT amps.
To date i have bought no less than 5 pairs of his regulators with the initial implementation in my troubled customized built for me type 26 line stage.
This line stage arrived at my door step with hum and rf noise issues and when I questioned the builder he told me it was normal and was beyond what could be done for the very hum my type 26 tube. I was not convinced!
Various postings and links on the very long type 26 preamp discussion also in this forum led me to read bout rod's regulators. I mean dc regulators aren't new so what's diff on this one, I really don't know then. I wrote to rod and got a reply with very favorable response as well as a no brainer purchase since it costs absolutely nothing compared to the ones being sold out there. I mean regulators are just regulators right as long as they reduced or eliminated hum?
No way! Various implementation of his regulators including on some of my newer projects has yielded a very interesting common feature bout rod's regs. I implemented them in high mu DHT triodes such as the EML20B as well as the ultra rare VT 64 /800 triode in a line stage application.
With these every implementation I find there is a surrealism bout the music where it really comes alive and draws you right into the music. Most will say all DHT's do that don't they, the fact they don't. I have in my collection amps build on AC heaters, unregulated DC heAters, regulated DC heaters using the more common current and voltage based regulators, and lastly Rod's units.
The difference, I remembered how when reading bout the practices of many amp builders or designers in the early days , AC heating was the preferred method. DC especially unregulated DC sucks the life out of the amps implementing them and I truly agree.
Rod told me then that even AC to him is bad and his regs will render them pale in comparison. I was skeptical but went with the idea.
To date the amps and line stages I have implemented them are the most musical sounding I have ever heard, for those who has referred to his manual stating' Your DHT amplification will sound so good that you will want to listen to your system for long periods of time' I couldn't agree more!
Rod, only 2 nights ago I replaced the LM regulators originally used in my 26 line stage to yours,( yes I have been really lazy! ) for doing so requires implementation of individual heater Trafos and means total re configuration of my ps chassis. Current configuration is a pi filter clc with the second cap in the signal chassis.
Only this morning I fired it up and was absolutely floored by the difference, as mentioned above immediacy and dynamism to the music! I never heard the 26 pre sounding this good, my previous regs were making the music mellow and overly smooth in other words boring.
I could rave on about his regs for pages but I think you pretty much get an idea of what I like bout these regs. If most of us think that dc regs are there to remove hum, think again, they could and will redefine music played from your gear at a whole different level!
kudos to rod on his awesome design I can only imagine with better component used what great improvements it could bring but anybody building a DHT gear should know that these are a bargain for the gobs improvement in sonic presentation.
Will post photos of my gear when I get a chance!
Are the Hammond dual bobbin transformers suitable? (Is there a difference between dual bobbin and split bobbin?)
http://www.hammondmfg.com/pdf/5c0018-19.pdf
Also I remember we need a regulator for each tube, but is it OK to use one transformer per two regs/tubes?
http://www.hammondmfg.com/pdf/5c0018-19.pdf
Also I remember we need a regulator for each tube, but is it OK to use one transformer per two regs/tubes?
HI
Can you tell me how you made those resistors?
You have entered one psu for one tube or something else?
I have a audiorome SET amplifier with 813's tubes(also has 12au7 and 6sn7)
In witch tubes do you think that I will expect to hear better things?
Tnank you Manolis
and a tip (i read in an electronics magazine of the 60's) is: on for example an ECC83 you can balance the tube halves completely this way without adjusting the circuitry itself. Their application: tube volt meter.
Has anyone seen a source for an R-core suitable for a GM70 filament application?
Hi Rod... I am using your filament supply now with my 45 DHT: that's clearly an improvement over AC supply...
I agree also with Petavgeris : the plastic 220 nF capacitor is a dead end, a good capacitor (I use MKP sprague orange drop) is clearly an improvement . Replacement of R1 by two cheap carbon: again, one step higher. Well , in this tweak session, I give 66% to the capacitor, 33% for the resistor... Next step: Duelund for the resistor, a Jupiter for the capacitor...
I agree also with Petavgeris : the plastic 220 nF capacitor is a dead end, a good capacitor (I use MKP sprague orange drop) is clearly an improvement . Replacement of R1 by two cheap carbon: again, one step higher. Well , in this tweak session, I give 66% to the capacitor, 33% for the resistor... Next step: Duelund for the resistor, a Jupiter for the capacitor...