DIY Audio Projects Forum

Recently I built a couple of 6L6GC-12AT7 SE UL power amps in a wine box. But there was no way I could get filament and HT trannies to fit and work in the wine box. I was using the step-down tranni to step down tranni but backwards method to generate HT. You need big trannies even for relatively low current draw in your power tubes (in this case 6L6GCs). What happens in the step up stage the load causes a lot of loss in the second tranni. The first tranni supplies 12V (the tranni is 24012V 10A) for the filament voltage and this is easily handled by 160VAC toriods. But the step tranni has too many losses and where voltage should be 240V our of the primary it is a little lower.

All this is fine if all you require after rectification and filtering under 300V but no good if you need 400V++. You could have the second step up tranni a 240-9V (160VAC). This would greatly boost the AC to be rectified to quite a high HT with good current handling. EL34s, KT88s etc. would work well here.

What I am proposing is a tranni box with two trannies to provide filament current and HT volts. Two step down trannies back to back. Mount them in a box with fuses and earthing and a switch and provide two XLR sockets at the back. Build a few different amps with rectification and filtering on board. The amps could have flying leads to connect to the tranni box, or as the tranni box, have XLR sockets on you amps with universal leads. Lead length and style could be customised (more on another post). Male sockets for the filament voltage and female for the HT. And leads to suite.

This would allow you to flip between amps but save on cost by using the one tranni box. Now lets take this concept a little further. Why not
have basic rectification and the first stage of filtering in the tranni box. Then you only need second stage filtering-storage and voltage limiting in the amp.

So now you have an number of tube amps with XLR sockets to receive filament voltage (AC) and HT pre-rectified and first stage filtered voltage. Inside the amp enclosure the filament voltage is wired direct where the HT voltage is limited and second stage filtered.

A short note about what I have found with my 6L6GC power and and SS rectification. Not only SS rectification but 850uf of filtering. The amp is exceptionally nice to listen to. Is it the fact that I'm using SS rectification or-and massive (for tube amps) filtering storage or is it the two tubes in use? Comments welcome.

_________________
Projects:"retro2308" - chip based headphone amp | ”Calibre 834” - tube phono MM preamp | ”najah” - Raw 180W Tripath Class D power amp | "Icon" - Shuguang CV-181Z preamp

Hi,

Years ago people used modular design with one set of output transformers which meant instead of having lots of cheap output Tx they just bought one set of expensive tangos etc and plugged the amp into them. This equipment also used modular power Tx and chokes for the same reason. However it never became popular because one amp tended to become the main work horse.

Really what you are suggesting is no different from a test PSU. The only thing to be aware of is the cable connections or exceeding the voltage rating of cables between modules.

It was done with preamps like the prophet as well as power amps:
https://www.google.co.uk/search?q=tube+ ... ZCjXumGvwM:

Vintage amps powered the preamp from the power amp using the power amp as the supply. However the power amp supply can modulate the pre supply due to sag.

SS power supplies are great with PP circuits and they don't need a choke current delivery on transient is also better. The snubbing can be an issue with SE.
Just my thoughts but the PSU has a major impact on the "feel" and sound of an amp. A bit like AC vs DC heaters or battery PSU.
SMPS can be just as good if not better! They can deliver current fast..

Its interesting that some people insist that voltage doubler supplies are the reason for some of the best vintage amps being so good.
I cannot really comment because I haven't experimented.

It depends on how you feel about the PSU having a sonic impact or not. If one size fit all then why so many PSU variations?

Regards
M. Gregg

_________________
What is the sound of one hand clapping?

Where this idea came from was with my latest amp, Silver Supreme, a 6L6 power amp in a wooden wine gift box. Using the one external tranni PS I could have a number of these amps with different tubes and amp roll them with the PS. Also I thinking of having a DC HT and filament outlet on the back to power any tube preamps. Because I build so much and love to try different tubes, interstage caps etc. this is the perfect way to cut build time and cost.

I couldn't imagine HT sag with such a robust PS and 850uf of storage caps. Most tube preamps draw very little current. I just see, for people like us, this being such a winner.

I should add that I agree the PS can influence the final sound. Only a very few of my tube amps have SS rectification but this amp sounds so good I'm wondering if SS rectification is not the best ways to go. And add to that heaps of filter and storage caps you have a dead quiet very robust PS all for a $1.50 1A rec. bridge.

_________________
Projects:"retro2308" - chip based headphone amp | ”Calibre 834” - tube phono MM preamp | ”najah” - Raw 180W Tripath Class D power amp | "Icon" - Shuguang CV-181Z preamp

Regards the sag problem,

If you think about a standard power amp it has a pre stage, driver stage, and power stage.
The pre stage PSU is connected to the power stage, so if the power stage uses transient current it can pull the pre stage PSU down with it. One way around this is to diode feed the pre stage PSU so it can only be fed from the output PSU but the output PSU cannot draw power from the pre stage.

With PP amps the ripple cancelation makes the use of SS rectifiers and not using chokes in the PSU viable.
However in SE amps the PSU seems more critical.
A choke in a PSU is always a problem with restricted current delivery during transient dependant on the capacitance after the choke being able to deliver pulses of current. The problem is then thought to be moot, however as with all things that's just half the story because the PSU must now recharge via the choke before the next transient.

Again in the past it was thought that over sized PSU smoothing caused SAG due to recharge times after transient.
So the problem with SS rectifiers also exists with diode type and commutation pulses. (Which can sound interesting with what appears to be better treble response). This is where pulse filters can be used, its enough to drive you nuts.

This is where people say choke PSU's sound different to resistor based types, and that both sound different with SS rectification.
This is where its possible to use HF parts in what appears to be a LF supply because the PSU has to deal with commutation pulses, radio noise and back EMF from the output Tx. This can be heard with OTL amps that don't have an OP Tx.
The removal of Back EMF seems to remove the warm woolly sound that is sometimes linked to tube sound.
The effect of the back EMF is to try to shut the tube down, worst case scenario is OP tube failure often linked to toroidal type OPTx's.
Years ago Velleman had this problem with the K4000 and used transient suppressors across the OP Tx's I think it was a couple of 440v VDR units shown here:
(Top left)
I think they were used in series from one anode connection to the other on PP to prevent a possible short burning out the OP Tx.
Slight digression but interesting.


Regards
M. Gregg

_________________
What is the sound of one hand clapping?

Hi, I am a believer in the PS being related to the sound of an amplifier. I consider it at least as important as the amp itself. I favor SS rectification with rather large filter caps. All must be bypassed as well. Several stages of filtering is a must for the driver and any preamp stages. Output stages are in my designs always PP and always class A. That maximizes the reduction of PS noise and being essentially constant in current demand. Low impedance for the return path of the output stage is also important. In addition to the large filter caps I normally use a large value poly right at the input to the output trannies. Another thing I do is select gain stages that have inherent PS hum and noise rejection. All together these features make the amps quiet and have fast transient response.

Now back to the topic. I don't favor separated power supplies for a number of reasons. First I am not thrilled with having 450 volts or so running around between pieces of equipment. Second the connectors can be problems. What happens if one is disconnected with the power on or the reverse, just plugged in with the power on. Where to put the power switch was an issue as well. If you put it with the amp then another set of wires need to go between the parts. More chance for problems and noise generation. The wires them selves can both radiate (depending on the percent of filtering done in the PS vs main amp) or pick up noise. The filtering can be problematic as well. My experience with such designs several years ago indicated that to achieve the same level of hum and noise filtering required one additional stage of filtering.

All said it can be done, I just personally don't see a value in it for amps. For lowest noise and hum ....possibly in phono preamps. But I can get -90db there with them both in the same chassis.

Good listening
Bruce

_________________
Some of my DIY Tube Amplifier Projects:

• PoddWatt: 5751 / EL84 Push-Pull DIY Tube Amplifier Project
• OddBlocks: KT88 Push-Pull DIY Tube Amplifiers (Mono Blocks)
• ForeWatt: ECC802S (12AU7 / ECC82) DIY Tube Preamplifier Project
• GrooveWatt: Tube (Valve) RIAA Phono Preamplifier Project

This thread gives me an idea of a universal design supply with switchable rectifiers (SS and tube) and filter capacitance. Thanks!

Edcor "Tinker Box" for the tranny would be called for, too.

_________________
-= Gregg =-
* Ratings are for transistors - tubes have guidelines*
Home: GeeK ZonE
Work: Classic Valve Design

If you had rectification and first stage of filtering in the external shared PS you could have SS or tube rectification. I'm trying not to make the external PS too big. By just passing the regenerated HT AC and filament voltage to the amp and the amp accomodating the rectification and filtering both remain a reasonable size.

At this point I'm just thinking of this idea as a compliment to a number of wine box amps. And of course you could have matching tube preamps or a integrated tube amp, even better.

HT sagging in the driver stage should not occurs as there is two 5K resistors (one per channel) feeding the limited HT (from 295V to 280V) to the tubes. At the end of each limiting resistor is a 47uf 450V cap. So HT should be steady at the driver tube (12AT7).

_________________
Projects:"retro2308" - chip based headphone amp | ”Calibre 834” - tube phono MM preamp | ”najah” - Raw 180W Tripath Class D power amp | "Icon" - Shuguang CV-181Z preamp

I have been toying with the idea of using direct mains to provide HT. I have tried it and it doesnt work. In fact you could take out the rest of your gear if used.

I used SS rectification and filtering. When I meter between the negative of the PS and my home earth there is 120V the difference. My mains voltage is 240V so it make sense that I may see a potential difference of 120V.

The rest of my gear has the ground side of its inputs and outputs connected to earth. If I plug in an amp with direct mains generated HT it could take out the rest of the gear.

I haven't tried tube rectification with direct mains yet. Until I do I'm staying with isolation methods for generating HT.

_________________
Projects:"retro2308" - chip based headphone amp | ”Calibre 834” - tube phono MM preamp | ”najah” - Raw 180W Tripath Class D power amp | "Icon" - Shuguang CV-181Z preamp

I hate to say this but direct mains is just asking for a fatality.

It was used in old gear and that is also very dangerous.
OK its a cost/weight saving but you end up with a floating chassis on mains equipment.
There should not be any connection between Earth and one side of a non isolated supply<<and it also makes RCD's fail.
So all the control knobs etc are potential death traps.
In old equipment the chassis was inside a wood enclosure with Bakelite knobs to isolate the control pot shafts and switches.
But it didn't take into account the women with wet hands just turning it up while washing the dishes.
With only two wire mains cord and plugs that could be rotated or wired up incorrectly the chance of a live to chassis connection is inevitable. With series heater chains if one heater goes open circuit the whole lot rises to mains potential.

Direct mains gives no limitation of fault current either.
Also if you do it you are directly liable for anyone who may come into contact with it, even if it's accidental.
You have to remember that mains voltage rectified or voltage doubled has no limitation on current (A fuse doesn't limit current)
You may if lucky survive contact with HT on an isolation transformer (capacitance storage dependant)
But the chance of survival on stepped up mains is much lower.

If you earth the chassis and use a separate electronic ground that is even worse, because you can be in contact with earth as you try to disconnect a phono plug which is at electronic ground!

In industry its very common to see floating electronic ground even up at thousands of volts but its a very specialised field and people are trained to deal with it. Years ago it was sort of custom and practice to have the earth removed on a scope so you didn't blow the equipment up with a floating electronic ground but its frowned on now and you would probably loose your job.
Enter the double insulation..phenomena.

Regards
M. Gregg

_________________
What is the sound of one hand clapping?

Well for me the trying is over. I'm now using double trannies and it is working well.

_________________
Projects:"retro2308" - chip based headphone amp | ”Calibre 834” - tube phono MM preamp | ”najah” - Raw 180W Tripath Class D power amp | "Icon" - Shuguang CV-181Z preamp