- Posted by cheap-Jack (A) on May 08, 2007 at 08:13:16
In Reply to: Re: External filament supply for 26 pre..? posted by wpod on May 8, 2007 at 03:48:52:
Hi.Go www.sayal.com where I got my SLA cells from.
c-J
|
Home
/ FAQ
/ News Classifieds / Events |
Audio Asylum Thread Printer |
Get a view of an entire thread on one page |
69.156.196.162
In Reply to: Re: External filament supply for 26 pre..? posted by wpod on May 8, 2007 at 03:48:52:
Hi.Go www.sayal.com where I got my SLA cells from.
c-J
Whta are 'SLA' cells? I have never heard the term.
...usually contain a gelled electrolyte and so can usually be used in any position compared to liquid electrolyte as in automobile battery. The SLA has some advantages over the common NiCad but low weight isn't one of them.
.
A SLA with a wide bandwidth series inductor?dave
Hi.Battery, dry or wet, is a very complex AC network by itself, techncally much worse than a capacitor.
To provide a bypass for the return signals, I installed a 5uF250VAC fan-motor cap, & a 333nF250VAC PP box film cap parallel to the battry O/P terminals. Also 2uF100V PE box film cap at the heater
DC supply I/P in the phonostage side.If the idea of addding a "wideband" series inductor is to remove the RF noises emitted from the SM charging board, then it would be needed.
As when the SLA cell is draining to power the tube heaters, the SMPS board is swithced off & electrically totally disconnected from the battery circuit. Of course, I already add RF ferrous suppressor ring onto the phonostage power cord as well as the DC O/P power cord of the batttery power pack as a RF safeguard.
c-J
the last thing you want to do is bypass the battery. since AC signal (your music) will appear on the + terminal it has two choices to get to audio ground. through the DCR of the filament or through the impedance of the filament supply. The wideband series choke prevents the music from using the battery as a path to ground.dave
Hi.Do I still need a wideband choke?
c-J
If you want to go down the SLA road you're going to need big batteries and efficient chargers - another poster suggested a 30 which is a fraction of the consumption and a very good DHT - one of a number of DHTs originally intended for battery use, which are typically 60mA to 120mA. But I will say that in a test of several different filament supplies, the batteries were not clear winners as one might expect, so don't expect miracles. And do expect to do a lot of recharging batteries rather than having a build-and-forget solution.
Hi.If you want PURE DC to power the tube filaments, what can be better than a rechargeable cell like SLA, cheap to own & no hassles to run.
I am using one for my phonostage heaters (0.6A6.3V) powered by a small 5AH cell, rechargeable by a switch-mode PS. The whole outboard power supply pack, comprising the SLA cell, the SMPS board & charge/discharge swithces/LED indicators/voltage check-points, etc, is DIY built in a small cheapie plastic equipment box (7"x2.75"x4.5"). It looks cool & neat, sitting beside my phonostage/linestage.
This is my "build-and-forget" pure DC supply solution for my phonostage tube heaters. No "miracles" if you know the way.
For 1A or larger current loads, you can go for 7AH or high rating.
One cell can do it all.c-J
PS: Any battery, dry or wet, is a complex AC network itself. Something got to be done to make it sound good. SLA cells are no exceptions.
'This is my "build-and-forget" pure DC supply solution for my phonostage tube heaters.'Really ? It seems that you built it then forgot a battery does not maintain it's charge . I bet a 5AH cell only has around four hours use , less if you've purchased cheap batteries which have past their shelf life date . How do you keep the heaters within +/- 5% of 6.3 volt for the heaters ? A much better solution would be to use an extra SLA in series and a CCS between the heater supply and the battery . This would maintain the heater supply at optimal conditions , even when the batteries begin to drain .
HI.Being a very very busy guy, I seldom get the luxury of 5-hour (!!) session for my vinyls.
BTW, you ever touch vinyls? If not yet, don't try it, amen!
Here is 'routine' rituals I go through to play a vinyl, put the LP on, secure it with the disc stabilizer, inspect & clean the stylus, check the DYNAMIC balance of the cartridge/stylus, wet the disc with pure distilled water (ad I always play wet), check the motor speed again, then lower the cartridge to the right track dead-slow manaully using the turntable hydraulic lever..........
But before I start these sorta painful rituals, everytime I check the cell voltage first, monitored by a dedicated voltage digital meter always inserted to the voltage check points installed on the power pack housing. I push the quick-charge button to start charging the cell if found drop a bit too low. This is always done before I start to exercise my time consuming LP routine rituals.
So far so good. Well, it worths the "pain" considering the joy of
musical enjoyment & sorta being-there engagement that only vinyl can offer.c-J
PS: Your CCS suggestion is correct, technically. But as a vinyl fan, I try to refrain from any SS sonic pollution after going such a long long way.
Neato, Jack. I've never heard of this being done.I've thought about doing this before, but didn't because I thought that the water would trap more airborne dust and cake it to the record's surface.
It really sounds like a good idea with many benefits (noise reduction and heat dispersal being the big ones) but I have a concern: does the viscosity (albeit very low viscosity) of the water affect in any noticable way the cantilever's recovery time? I realize that the resistance of pure water is going to be very low, especially when the force is being exerted by a recovering (returning to zero on axis) low-compliance cartridge, but it still seems that there would be a noticable affect...probably an audibly positive one.
I like the idea. I'm going to try it!
...esp when cost is factored in.
SLA cells are indeed one way to go, but I didn't find they sounded better than a good current source filament supply when I tested a few different supply types. A supply consisting of two voltege regs followed by a common mode choke was pretty good. You'll need a heatsink with the 26 because the filaments are 1 amp, and the size will depend on the filament supply and the voltage dropped by the devices you use, but be prepared for something of a decent size.
This 26 is very similar to 27 I built which is outstanding w/ my 300B PP amp open baffle setup.Simple cap coupled jobbie that really sounds good.I am using all wood chassis w/ large copper ground buss as I did with the 27. I see my plate voltage is much lower than other schematic's I've seen,..? The simple filment supply shown is what I will try 1st w/ even more uf, about 300k per filament, might add a choke to filament supply also.?
Hi!nice simple design. What is the input level you expect the 26's grid will see ? Nominal CD level would drive the grid positive already.
If you stick with this op point, may I suggest to give 'filament bias' a try ? Connect the filament supply to one filament pin and the other filament pin to ground through a 1.5 Ohm resistor. This will give you the same grid bias without the need to resort to a battery. However in this arrangement the filament supply must be very quiet. I'd recommend a cutrrent source, or better yet a choke. Make it LCL, if you have the iron available.
Best regars
Thomas
Hi.an effective decoupling filter is added downstream of the regulated PS, which is crucial for phonostage PS.
c-J
Nope they don't sound at all good this way decoupling or not ! About the last thing you need is decoupling , what is required is isolation . Chip regs (adjustable types) are best used as current regs but everytime I find it better using LCL . Also for critical filament (not heater) applications , current reg connected IC's can be noisy . Filament supplies are critical to good sound of DHT's , much more so than with heaters but practices for filaments can be implemented to good effect . A typical cap input supply filament followed by an LT1085 type chip can much be improved by changing the supply to choke input and improved once again by removing the chip and replacing it with a choke . Not all chip regs are equal either . Chips made using legacy technology such as LM317 sound awful and are noisy as CCS , LM108x and LT108x series are better but not as good as a choke . For phono stage use which you mention ,(which normally use indirectly heated tubes) balanced supplies provide optimum performance rather than the type of regulator chip
Hi.Or it just your own way of describing it?
c-J
it's LCL and I just have ! CLC does not isolate the filament from the last cap in the filament supply . The final choke isolates the smoothing cap from the signal which passes through the filament . For reasons which I do not understand , a solid state CCS , with it's huge impedence sounds inferior to a passive LCL supplied filament which relies on the reactance of the final choke . This is a well known practice of DHT afficionados . You obviously have never tried or even considered this
For reasons which I do not understand , a solid state CCS , with it's huge impedence sounds inferior to a passive LCL supplied filamentNot all CCS are born equal. Which one(s) have you tried?
DN2540 and LM1084 for 250mA fils running from a battery supply , had problems with the drop across the DN2540 . LM1084 and LT1085 for currents above that . I'm yet to try mosfets . So far I've found that the CCS reg followed by a cap to 0V and a choke feeding the filament provides the best sound when a chip reg is in the chain . LM317 sound horrible , are noisy and the filament needs to be bypassed with a cap . The heavy , bulky LCL approach wins hands down every time ;)
Thanks for the info. The LM317 is indeed horror. Can I ask what kind of choke you use and where do you get them from? Thanks.Arend-Jan
LM317 is fine for heaters ;) Chokes are mainly from radio rallies , where else could you find a 10mH 10A choke for example ? So far I've been lucky and been able to pair up many of the singles I've found . Ebay is also a good place to find stuff , although postage price can make some deals unattractive . I'm beginning to think that rewinding and gapping mains transformers may be a worthwhile alternative . Chokes in my linestage are 280mH 2.7A (more like 100mH at the 1A I'm using these at) from Display Electronics in the UK
So what would be a good LCL filament supply for a 26..? Just wanting to learn.
what filament chokes do you have in stock or access to ? DCR of a hot 26 filament is 1.5 ohms , finding an isolation choke with sufficient reactance should be fairly easy . Just treat the LC part as you would a choke input supply . My main advice regarding implementing this is to use an outboard supply , keep all the iron well spaced and well away from the valves , use schottkeys for rectification . You could always go the silly route and use a pair of CV5 MV rectifiers instead of the schottkeys if you want to avoid solid state altogether :)
The MV route has me intrigued, I am a silly type guy anyway. So how ould one implemiment the cv5 MV..? & what is the equivalent # of a CV5, I really like running the vr75's for reg, so why not go MV..even if a hassle, I care not.. Chokes I have on hand are 1- Hammond 193L-5H @ 300ma, 2- 17H with unknown ma @ 300ohm DCR made by National Radio Company & are potted. I have access to any of the new production Hammonds & can source what ever might be needed.I also have the 2- 9H 40ma chokes that are in my 27's PS that I could use also.
You need low inductance , low DCR , high current chokes rather than the other way round . CV5 is a beast , 4 volt 10 amp filament , mogul base and are about the size of an 845 . I have a pile of CV5 going spare if you're keen ;) I have never used an MV for filaments , although a pair should easily light up a 26 . Definitely silly 'silly' , 80 watts + CV5 volt drop all to light up 1.5V @ 1A :) Maybe a pair of 866A would be a more conservative option ? (hehehe)
