• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

6V6 line preamp

diyAudio Chief Moderator
Joined 2002
Paid Member
Feasible then...

- The info that 6V6 is already used in line circuits with good results is positive. Thanks!

- Here is a basic circuit I came to graphically. Does anybody have a 6V6 model in some sim soft package? I sadly don't have it in my sims.

- If yes, lets see if the operating point and B+ are nice or there is better.


  • 6v6line.jpg
    61.5 KB · Views: 26,313
Official Court Jester
Joined 2003
Paid Member
Re: Feasible then...

salas said:
- The info that 6V6 is already used in line circuits with good results is positive. Thanks!

- Here is a basic circuit I came to graphically. Does anybody have a 6V6 model in some sim soft package? I sadly don't have it in my sims.

- If yes, lets see if the operating point and B+ are nice or there is better.

these simple tube circuit deserves that you learn graph-ruler-calculator routine.

you don't need sim for that ; or you'll never learn how tube works
6V6 models

Does anybody have a 6V6 model in some sim soft package?

Try these models. They work on LTSpice.
*  6V6 tube has same curves as 6AQ5
*   capacitances are slightly different
+ PARAMS: Vn=4.683375986 Exd=1.498943398 Kd=5.985681E-04 Mua=135.8092472 Mug=0.792313597 Mus=7.857142857 Kp=7.41E-04
+ Ks=0.7455 Ex=1.429346869
+ Cgk=9.0P Cgs=8.0P Cga=0.7P Cak=7.5P
Eat at 0 VALUE={limit(0.636*ATAN(V(A,K)/{Vn}),0,1e6)} ; arctangent shaping
Eme me 0 VALUE={PWR(LIMIT(V(A,K),0,2000),{Exd})*{Kd}} ; the diode line
Egs gs1 0 VALUE={LIMIT(V(A,K)/{Mua}+V(S,K)/{Mus}+V(G,K)/{Mug},0,1E6)}  ;the basic voltage
Egs2 gs2 0 VALUE={PWRS(V(gs1),{Ex})*{Kp}}  ;raise to the power and mult by perveance
Ecath cc 0 VALUE={LIMIT(V(gs2)*V(at),0,V(me))}  ;shaped per Eat and the diode line
* Calculate anode current
Ga A K VALUE={V(cc)}
* Calculate screen current
Escrn sc 0 VALUE={{Ks}*V(gs2)*(1.1-V(at))}; reverse arctangent shaping
Gs S K VALUE={V(sc)}
* Grid current

* Capacitances
C1 G K {Cgk}
C2 G S {Cgs}
C3 G A {Cga}
C4 A K {Cak}

* 6V6
* 28 Nov 2004
* new model based on measurements
*  parameters revised 29 Nov 2004
.subckt 6V6GT A S G K
XV1 A S G K Pentode2
+ PARAMS: Ex=1.41031468452635 Mug=11.7888046293233 Mus=36.3212284562041 Kg1=1084.30402294001 Kp=36.0452435978759 
+ Kvb=899.32403046434 Go=-0.467417459629955 Kd=0.000106867654972877 Exd=1.26175320982325 Exsd=0.425953018509995 
+ C1=-2.85604234699625 A1=-5.68835011136307E-05 B1=3.49035244298882 X1=0.015108363529097 Y1=0.00926342558705693 Sr=0.198544592645026
+ Cgk=9P Cgs=5P Cga=0.7P Cak=7.5P
.ends 6V6GT

*  This model is for pentodes with remote cutoff grid characteristics and with
*  knees that blend into a smooth line, ie a diode line, typical of Beam type tubes.
*  this model therefore uses Korens method for total current, and diode lines.
*    default parameters are for the 6V6
*  RCM  2004 November 28
*      2004 Nov 29  revised capture ratio formula to correct problem when Vg > Vp or Vg > Vs
.Subckt Pentode2 A S G K 
+ Params: Ex=1.42596448566733 Mug=11.7822188645457 Mus=27.5137445137459 Kg1=1447.15506010676 Kp=36.3066403677756
+ Kvb=756.918578086785 Go=-0.557407825110845 Kd=0.000103636138847149 Exd=1.26335466904983 Exsd=0.431363489073363
+ C1= -0.940657929992395  A1=-0.000127738832949614 B1=1.72299135911867 X1=0.0345083152949822   Y1= 0.0177418846473534  Sr=0.0471284218217744
+ Cgk=9P Cgs=5P Cga=0.7P Cak=7.5P

* calculate the total space current based on Koren's triode equations,
* but with an equivalent triode plate volatage equal to the pentode Vs + Vp/Mus
En1	n1 	0	VALUE={(V(A,K)/Mus+V(S,K))/Kp*LN(1+EXP(Kp*(1/Mug+(V(G,K)-Go)/SQRT(Kvb+(V(A,K)/Mus+V(S,K))**2))))}
Esc	sc	0	VALUE={ (PWR(V(n1),Ex)+PWRS(V(n1), Ex))/(Kg1) } 

* calculate the plate current capture ratio
Ecr	cr	0	Value={If(V(A,K)<=0,0,C1+A1*V(A,K)+B1*((limit(V(A,G),1,1e6)**X1)/(limit(V(S,G),1,1e6)**Y1)))}

* the tentative plate current is total space current multiplied by capture ratio
Ept	pt	0	Value={V(sc)*V(cr)}

* calculate the diode line current
Ed	d	0	value={Kd*V(A,K)**Exd*V(S,K)**Exsd}

* the actual Plate current is the lesser of pt and d
Ep	p	0	value={Min(V(d),V(pt))}
Ga	A	K	value={V(p)}

* calculate the screen current capture ratio
* if Vg <= 0 then the screen gets all the remaining current, otherwise the grid gets some
Escr	scr	0	value={Limit(if(V(G,K)>0,1-Sr*V(G,K),1),0,1)}

* screen current is anything not captured by the plate, multiplied by screen capture ratio
Esg	sg	0	value={(V(sc)-V(p))*V(scr)}
Gs	S	K	value={V(sg)}

* grid current is anything not captured by the plate or the screen
Gg	G	K	value={V(sc)-v(p)-V(sg)}

* interelectrode capacitances
C1 G K {Cgk}
C2 G S {Cgs}
C3 G A {Cga}
C4 A K {Cak}
.ENDS Pentode2

I have a line amp that uses (or used, more in a minute) the 6AQ5, which is a miniature 7-pin 6V6, and it worked a treat:


Except that I had a terrific problem with microphonicity in the 6AQ5s. They have cylindrical plates which ring like a bell, most of them. I now use a different tube, but I haven't noticed any significant microphonicity in 6V6s, which I use as power tubes, so you are the victim of a Good Idea. I have a more complete project description on my site:

The Project

If you have any questions, ask away.


diyAudio Chief Moderator
Joined 2002
Paid Member
Poin(t)z taken!

Hello Poinz

-Nice to have seen your gear favoring highly on 6moons. Congratulatios!


-I had those RCAs and I was thinking of building your Musical Machine maybe with an E80CC first stage twist a long ago, but that idea fell victim of needing the power that my EL34 / KT88 amp with AN Trans 185s gives, I had developed it to a satisfying degree by then, when my speakers got 4 more woofers added. They are 94dB but 4-6 Ohm now.
Abandoned project's nice tubes got to play music through something else I thought, and not to idle in drawers.
Somehow I felt that a 6V6 triode strapped will be gutsy and musical as a line amp with 350V B+.
I simulated the rough circuits I posted in meantime (thanks again for the models Canada) and I saw that they seem mainly ok. Only change is I can use a 560R instead of a 500R to go slightly more linear. Thats fortunate because 560R is standard value and I will get 2W Kiwame for that. For the CC circuit 2x10K 5W Kiwame in parallel will do (3.7W idle dissipation needs 2-3X part rating not to become brown and fail soon if of 5W rating). Also the 180R can be 100R for slightly more gain (100R-220R totally ok basically). That gain will be about 7X cathode bypassed for CC and maybe 5.5 for SRPP. The tubes shall idle at about 5.5-6W which is conservative for life time but not skinny. Any 100uf to 1000uf 25V Ck cap will do. I like Silmic and BGN. I prefer 1000uf bcs the lower the corner the less the cap's THD. I dont like the sound of local feedback. I even find a little loop NFB better sounding than comparable amount of LNFB, seriously!:xeye:
Will take me some time to build from scratch, but an idea is to get some noval to octal adapters and use my ECC82 preamp rewired for testing. It sports a 360V B+ PSU with 5R4GY RCA rectifier tube , 3x 20H chokes, 2x current source heater supplies (have to modify 2 resistors only for 0.45A per tube) has selectors, RCAs, pot, etc. Can I string 2x 6V6s per channel and feed 0.7A in the SRPP case? Don't tell me they have some allergy for series heaters please! I JUST HOPE the 6V6s will not turn out ringing in a line stage! Especially after I will probably have my preamp gutted for their charms!

Nosy questions

-Should I forgo the CC circuit? What do you think? I see you ended up SRPP. Is the CC veiled? Why you went so low on B+? I can see that nice linear niche to the left plate curves chart corner, but 2V auto bias isn't precarious with CD sources of (at least) 5.6V p-p output? Also the RCA manual gives cathode to heater 200V max and you have (175/2)-6.3=81.2V C to H (only) on upper tube, so you float your heaters for noise reasons mainly?
diyAudio Chief Moderator
Joined 2002
Paid Member
Thanks Bud!

BudP answered my email very kindly.

He confirmed use of 6V6 as a current sourced CF (driving an SE to balanced output conversion transformer) after a 6G6 CC stage. He said it sounds much better than with smaller tubes. He uses 16V cathode op. point.
I chose 15V (I hope it proves there about in real test). So all lights are green.
I think that the big broblem for this project is the microphnicity, i have build a pre with 6v6 tubes 3-4 years ago.. and as i remember i rejected many 6v6s for this reason..only two Bendix 5992 [6v6] and two RCA black plate was enough antimicrophonic and suitable for this case..:)
diyAudio Chief Moderator
Joined 2002
Paid Member
I do have 4 RCA black plate smoked glass 6V6s, testing very strong. Lets cross fingers! I will test soon on a donor 6SN7 octal pre with tube regulated PSU. So no gutting of my pre yet...
How was the performance?
  • Like
Reactions: 1 user

The perfomance was excellent,,big sound strong and also soft,,I think that the output power tubes as preamplifier have better sound than the normal small tubes [forgetting however the microphonicity and heater noises] .. I attach a pic from an extraordinary preamplifier with 5998 triode tube :)


  • 5998 line.jpg
    5998 line.jpg
    42.3 KB · Views: 7,872
Not 6V6, but I am for the moment building a preamp based on EL84. The design goals are balanced output, low output impedance and ~10dB gain. I will use it together with a solid state power amp ("ICE power", class D mono blocks).


There will be two chassis, a separate power supply. The preamp will look something like this:


Jan E Veiset
diyAudio Chief Moderator
Joined 2002
Paid Member
Initial test done!

Hello again! Rough test gave very good results. The regulated supply did not hold up to more than 260 VDC B+ so I used a 17mA ,11.2V cathode operating point. Wire wound sandcast plate 5W resistors, 680R 1W generic carbon rk and generic Matsushita ck. Basic white Mundorf (SCR) coupler. The common cathode circuit was tested.
Well...bigger 3D, no edge, more reverb and depth info, more musical detail than the 6SN7 cc&cf or SRPP, or grid leak cc&cf, or 12AU7 cc&cf that have been used previously in the system.
Piano sounds superb and big drums too. Lots of air and body but no flatness or edge nowhere. Voices more liquid and human, better defined with better front to back spatial resolution. Strings are velvety, basses huge and open with no pace loss. Very good drive, optimal gain. Treble stems out seamlessly from upper mid naturally, not like a distinct entity.
Microphonicity is confined to a glassy low level 'ding' sound when hitting the 6V6 glass with a fingernail or when turning the stepped attenuator located right next to it. No locking with the feedback from the speaker just 0.5m near. After half an hour some periodic 'hsst' sounding transients showed up accompanied with woofers travelling on right channel but I am suspicious of the small regulated supply 12B4 pass element getting taxed with heat. Maybe it was the right 6V6 heater aswell, but I doubt it bcs of the cone travel, looks like a DC shift. I will test with heavier supply in due time. Now its given away to be further auditioned and inspected in longer term in another system by a fellow DIYer. If it turns to behave well in the long run it will very probably get adopted as my premium pre with proper custom PSU, best parts and antimicrophonic box build (there is more performance to be expected given the compromised B+ and generic components first test). Will keep you posted.


  • 6v6bcrop.jpg
    80.4 KB · Views: 7,279