Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

12AX7, EL34 or 6L6 design wanted
12AX7, EL34 or 6L6 design wanted
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 4th September 2005, 04:51 PM   #11
amperex is offline amperex  United States
diyAudio Member
 
Join Date: Feb 2005
Location: MI
Default Don't start 'in a hole'

Interesting how hundreds of designs use less than best quality tube types- go figure. One has to consider all the handcraft work required to achieve a less than best sounding amp- no thanks! Also consider the DIY amp is hard to sell vs upgrading a vinatge amp.

Well known tube type great performers are Sylvania or Tung-Sol 6P5GT, RCA 6J5GT & Sylvania 6SN7GT. Power tubes vary a lot & one must consider power needed. For the most part, triode connected tetrodes perform best, but at 1/2 power. A regulated screen supply is inportant with tetrode connected tubes, however use a choke & poly or oil cap after that regulator. As for Ultra Linear, I consider that to mean Un Linear (stay away).

15-watts? Use PPP Sylvania 6V6GT triode connected or inexpensive $2 NOS Sylvania 5V6GT with a dropping resistor at filament (Radio Daze in the state of New York). Sweet sounding tubes.
See the 'poinz' design, but use a 6P5GT or a 6J5GT driving a Sylvania 6SN7GT or a pair of lower cost 6J5s for the phase splitter. BTW- The RCA metal 6J5 sound as good as the RCA glass 6J5GT. Not true of other metal 6J5s I listened to. I seen NOS on ebay for $2 each.

30-watts? triode connected Sovtek 6550 with rectangle shape plate holes, not the old stock inferior Sovtek 6550s with the round plate holes (please).

Hope this helps.
  Reply With Quote
Old 4th September 2005, 11:16 PM   #12
Johan Potgieter is offline Johan Potgieter  South Africa
diyAudio Member
 
Join Date: May 2005
Location: Pretoria, South Africa
Hi Fiat

Up to now nobody seems to have replied to your post#8, so let me try to help (I will explain simply and fully so this will be a little long.)

Firstly, the high frequency response will suffer with any appreciable input resistance. The main culprit is what we call Miller capacitance that is shunting your input. The highest input resistance with a 500K pot will be when the rotor is in the middle (resistance-wise). There you will have 250K to common in parallel with another 250K up top, also essentially in parallel with the first 250K, if the input source impedance is considered negligible for now. That brings us to 125K.

The main contributor to input capacitance is the 1.7 pF between anode and grid. This sounds low, but this capacitance is not to common, but to the anode, which for every volt across it from the grid side, has about 70V in the opposite direction on the other side from the anode (the amplification of the first tube will be about 70). The effect of this is that the grid "sees" an equivalent capacitor of 71 x 1.7 = 120pF. When this appears after the above 125K resistance, you will have a 3dB attenuation at 10KHz. If your source impedance is significant, this will be lower. There is also other internal capacitances acting here which I have ignored for simplicity sake.

You can now work out that at either end of the pot, the nett series resistance will be less and the frequency cut-off higher. Thus you will have worst attenuation of treble with the pot in the middle (resistance wise).

As an aside, this is why I am not fond of the high mu ECC83 as an input tube unless one can guarantee a low source impedance - and then not with a volume control in the way! Your best solution here is to use a 100K volume control if your input source is not shunted too much by that.

The expected output power will be about 8W into 8K transformer primary impedance. With only 250V applied the 6L6s are rather wasted. Once you get this to work, you can increase that to 450 - 500V with about 300 ohm common cathode resistor (10W). That should increase available output power to about 14W, depending on the output transformer.

Using the distributed load (UL) configuration can convert this circuit to give about 40W - a worth-while increase (at 450V h.t.). I must respectfully disagree with Amperex about the non-linearity - that is not what I have found or what my data shows (I have graphs from G.E.). UL allows one to mainatin about 80% of the advantages of triodes (low internal impedance and low distortion) but with about pentode output power. But perhaps that is for another day, if you want to start with the circuit as is.

Good luck!
  Reply With Quote
Old 5th September 2005, 01:33 AM   #13
amperex is offline amperex  United States
diyAudio Member
 
Join Date: Feb 2005
Location: MI
Default Most respectfully

Tetrodes respond best with a regulated screen as long as the regulator has some isolation from that regulator that can produce noise. Or, an option is a stiff seperate supply.

A better sounding amp simply uses a triode, but higher power triodes such as 845 are costly. A compromise is some tubes respond well to triode connected such as the 6V6GT.

UL has always produced the worst sounding amps of the lot & I will never go back to them. I consider UL a waste of time & money.

I like neutral sounding amps. The word 'warm sounding' just means audio distortion to me. Once one hears a fast & neutral amp, they can not go back to the mud.
  Reply With Quote
Old 5th September 2005, 01:34 PM   #14
Fiat1 is offline Fiat1
diyAudio Member
 
Join Date: Oct 2001
Thanks for your replies! Johan Potgieter, thanks for the great explination. I might increase the 250volts, since I have a schematic for a power supply using 2 off the shelf transformers conected back to back. They are both 160VA, with the first supplying 25VAC to the 30VAC center taped secondary winding of the second transfomer. At full load that will produce 150VAC which feeds a voltage double cct. This produces +480V. Or is there an easier way? If I could increase it to 14watts approx to start off with thats about what I am after. If all goes well then I will want to to increase the power to 40watts as sujested and possibly make my home theatre setup run of these.
  Reply With Quote
Old 5th September 2005, 10:18 PM   #15
Johan Potgieter is offline Johan Potgieter  South Africa
diyAudio Member
 
Join Date: May 2005
Location: Pretoria, South Africa
No fine, Amperex, I respect your views - and ears!

I would agree that used as a tetrode a regulated supply is required. Many "bit the dust" by feeding beam tetrode screens off the supply for the rest of the amplifier, where this was resistor-decoupled. I have even seen commercial designs that way.

I would also agree with the meaninglessness of terms like "warm" and "musical" and a host of others, however well intended. The latter sounds especially pious and seems to be something with which one can hardly disagree, but leaves one with the impression of an amplifier with the kind of harmonic content (addition) that makes the music sound "better". The customer can naturally buy with his money what he likes, but a lot of unnecessary ill-will resulted when the argument branched out to "A is better than B" instead of simply staying with "I (personally) prefer A to B".
  Reply With Quote
Old 7th September 2005, 09:05 PM   #16
aletheian is offline aletheian  United States
diyAudio Member
 
aletheian's Avatar
 
Join Date: Apr 2004
Location: Philly
Quote:
Originally posted by Johan Potgieter
Hi Fiat

Up to now nobody seems to have replied to your post#8, so let me try to help (I will explain simply and fully so this will be a little long.)

Firstly, the high frequency response will suffer with any appreciable input resistance. The main culprit is what we call Miller capacitance that is shunting your input. The highest input resistance with a 500K pot will be when the rotor is in the middle (resistance-wise). There you will have 250K to common in parallel with another 250K up top, also essentially in parallel with the first 250K, if the input source impedance is considered negligible for now. That brings us to 125K.

The main contributor to input capacitance is the 1.7 pF between anode and grid. This sounds low, but this capacitance is not to common, but to the anode, which for every volt across it from the grid side, has about 70V in the opposite direction on the other side from the anode (the amplification of the first tube will be about 70). The effect of this is that the grid "sees" an equivalent capacitor of 71 x 1.7 = 120pF. When this appears after the above 125K resistance, you will have a 3dB attenuation at 10KHz. If your source impedance is significant, this will be lower. There is also other internal capacitances acting here which I have ignored for simplicity sake.

You can now work out that at either end of the pot, the nett series resistance will be less and the frequency cut-off higher. Thus you will have worst attenuation of treble with the pot in the middle (resistance wise).

As an aside, this is why I am not fond of the high mu ECC83 as an input tube unless one can guarantee a low source impedance - and then not with a volume control in the way! Your best solution here is to use a 100K volume control if your input source is not shunted too much by that.

The expected output power will be about 8W into 8K transformer primary impedance. With only 250V applied the 6L6s are rather wasted. Once you get this to work, you can increase that to 450 - 500V with about 300 ohm common cathode resistor (10W). That should increase available output power to about 14W, depending on the output transformer.

Using the distributed load (UL) configuration can convert this circuit to give about 40W - a worth-while increase (at 450V h.t.). I must respectfully disagree with Amperex about the non-linearity - that is not what I have found or what my data shows (I have graphs from G.E.). UL allows one to mainatin about 80% of the advantages of triodes (low internal impedance and low distortion) but with about pentode output power. But perhaps that is for another day, if you want to start with the circuit as is.

Good luck!

Yeah...what he said... sorry, I have been away for a while. There are definately better ways to tweak the design... I was going for 'quick and dirty' with the lowest component count possible.

PEACE!
  Reply With Quote
Old 7th September 2005, 09:21 PM   #17
aletheian is offline aletheian  United States
diyAudio Member
 
aletheian's Avatar
 
Join Date: Apr 2004
Location: Philly
One other detail... this might be assumed, but I wasn't sure if this came across. If you increase to a higher plate voltage, then run the debate is to run it in Ultralinear or tetrode, which means you would have to change the grid configuration and potentially the power supply. new manufacture tubes in triode at 480V should work out, but it makes me nervous...
  Reply With Quote
Old 7th September 2005, 10:26 PM   #18
Johan Potgieter is offline Johan Potgieter  South Africa
diyAudio Member
 
Join Date: May 2005
Location: Pretoria, South Africa
Yah. . . . know the feeling. I am at present in the last stages of finishing a 120W amp using a nett 520V anode and 460V screen voltage (4 x 6L6GC in the Quad II type output stage). Having become used to transistors I approach carefully - hope no smoke and "spitzensparken". Folks may have noticed that I was looking for extra genuine GE - 6L6GCs on another thread; if unavailable (as it is beginning to look) I will be going to Svetlana or ...EH - then my turn to be nervous.
  Reply With Quote
Old 10th September 2005, 09:19 AM   #19
Fiat1 is offline Fiat1
diyAudio Member
 
Join Date: Oct 2001
So what do I need to modify if I want to increase the B+ volatge? Can I still use 12AX7's and a volatge of around 400volts? I ask this becuase the specs for a 12ax7 say a max plate voltage of 300volts? Or would a resisitor be used for their plates to drop the volatge? I am new to valve amps and their designs so bare with me, I am still doing reading on the subject. Thanks for all the help up to now!
  Reply With Quote
Old 10th September 2005, 11:28 AM   #20
Johan Potgieter is offline Johan Potgieter  South Africa
diyAudio Member
 
Join Date: May 2005
Location: Pretoria, South Africa
Firstly you can certainly use the ECC..s with a supply of 450V (seems to be a popular voltage for this sort of thing, probably because that is a popular rating for high voltage electrolytic caps).

One manual I have gives absolute max. for ECC83 as 600V, operating as 330V. The max. voltage rating is somewhat coupled to the dissipation and other factors - it does not mean it will start sparking inside at 340V! The thing not to exceed is the maximum anode dissipation (apart from keeping the heater voltage within spec.).

Then also notice that voltage amp. stages are decoupled from the main supply by say 22K (and 16UF/450V cap), making the voltage there no more than about 350V - 380V. Then you still have the anode load resistors. This is a longer story, but one tries to put the final anode voltage where the maximum required sine wave will just fit in with say about 20V above the G1 = 0V curve on the anode voltage/anode current load line.

Sorry that this comes to you rather disjointed - at which circuit are we now - if it is post #3 by Aletheian, perhaps I can hook up a practical circuit in a day or 2 (if he does not come first) and give you ball park values.
  Reply With Quote

Reply


12AX7, EL34 or 6L6 design wantedHide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Wanted. Simple EL34 Push Pull Design jack firebrace Tubes / Valves 51 29th July 2013 12:47 AM
Tubes for Sale - Link to List 12AX7,5AR4,6550,EL34 and More baileyler Vendor's Bazaar 3 22nd April 2008 01:20 PM
EL34 SE triode mode A class sch wanted trancy Tubes / Valves 14 10th January 2008 10:16 AM
PP EL34 circuit wanted Norris Wilson Tubes / Valves 3 9th August 2007 11:29 AM
6L6GC/EL34 SE Stereo Design bartr0n Tubes / Valves 25 16th September 2003 06:03 PM


New To Site? Need Help?

All times are GMT. The time now is 04:28 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 15.00%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio
Wiki