Can't decide which way to go...

General discussion area for tube amps.

Moderators: pompeiisneaks, Colossal

Post Reply
User avatar
WRC34
Posts: 169
Joined: Fri Dec 19, 2014 3:40 am
Location: United States
Contact:

Can't decide which way to go...

Post by WRC34 »

I'm waiting for a few parts to arrive before starting next build in earnest. I have re-wired and done some maintenance on a few amps but this will be the first amp I have built this year. Busy!

Chassis is punched & drilled - I'm using a dual input preamp setup that I found somewhere out there (good chance someone here either made the attached drawing or knows who did) where input 1 has it's own volume control and feeds into input 2 - meaning Input 2 is placed in the chain after V1 so Input 1 can be blended with V2 but Input 2 does not utilize V1 at all. I used this Input jack arrangement before on an amp that has a parallel 6SL7 as V1 and a 6SJ7 as V2 and it's cool/useful. Input 2 still breaks up a bit when cranked but is much cleaner than input 1. The values used on V2 (6SJ7) were fairly conservative so as to not exacerbate the potential microphonic issue these vintage pentodes seem to suffer from, yet it still sounded a little boxy when cranked with most 6SJ7s I have. V2 will go into a 5F2 style volume/tone control which will then go on to a 12AX7 long tailed pair into two 6L6GC/KT66/EL34/even 6V6 cathode biased output stage. Amp will have Solid State Rectifier w/large-ish filter caps.

Using a Classictone 40-18065 Trainwreck-style PT which has multiple secondary windings 300-260-0-260-300. Will use a large DPDT switch to choose between windings (obviously when powered off) to better suit the output tube type. This worked well on another build with their 40-18095 PT. Output transformer is an old "40 watt" laydown type, similar to what you'd find in a Leslie organ maybe? It says TRA-10-15 1005411 on the end cover. Tested through variac set to 120VAC input and it showed 5K primary. It's got 4/ 8/16 ohm secondaries. Using 3H choke for output tube screens as well. Transformers are all mounted on chassis, ready to go. Going to start with a shared 250R 25W cathode resistor bypassed w/a 100uf/100v cap for output tubes and take it from there.

Original thought was to wire up V1 with a DPDT series/parallel switch and V2 hard-wired parallel into the LTPI, but I cannot find a reasonable schematic out there for the switch that doesn't put DC on it and hence risk instability and/or arcing within the switch (or so I've been reading!) I have no real world experience with a series/parallel switch on a 12AX7, anyone have any tips? But now I've been thinking, maybe a pentode like EF86 in V1 could be an alternate plan. That is yet another preamp arrangement that I am turning up little to no info on. There's some info regarding an EF86 being used after a 12AX7 (see drawing!) and I have sort-of made this with the 6SL7 into 6SJ7 amp described above. Is it possible no one made an amp with an EF86 as V1 into a parallel (or series or single triode) 12AX7 which then goes on to the phase inverter? Will I be better off sticking to the original plan? Or maybe two parallel 12AX7 gain stages will be enough to drive the big bottles considering the minimal loss from lack of a real tone stack?

Not sure which direction to take this one in...any influence would be welcome!
You do not have the required permissions to view the files attached to this post.
ChopSauce
Posts: 1038
Joined: Wed Mar 15, 2017 12:08 pm
Location: So Paris, France

Re: Can't decide which way to go...

Post by ChopSauce »

So, you did already use any such V1, V2 arrangement with an EF86? I am curious about the component values you used for the EF86.

I found an EF86 as V2 - right after the usual first triode - sounds pretty disapointing, in terms of sound coloration. The EF86 was settled as per Merlin Blencowe's guidelines - that is at moderate (circa 80) gain. Another friend had similar observations. He suggests that the EF86 was put after 2 triodes and I am going to try that next. That or pushing the EF86 to failure-prone (higher) gains settings...

... I wouldn't bother wtih an EF86 I had to do it again, but maybe this is me having been missing something :?:

Also, the high output impedance of the EF86 appeals for some tone stack tuning, such as the use of a MOSFET follower, but you most likely know it better than I do
User avatar
WRC34
Posts: 169
Joined: Fri Dec 19, 2014 3:40 am
Location: United States
Contact:

Re: Can't decide which way to go...

Post by WRC34 »

Thanks for the reply Chop Sauce!

Funny you should mention that you weren't too impressed with the EF86 after a triode stage. I liked the 6SJ7 after parallel 6SL7 as described above, but it needs more tweaking. I also like EF86 alone into 12AX7 long tailed pair and straight into power amp. I built an amp like this w/EL34s and simple vol/tone controls and it kills. EF86 component values were very close to standard "Vox" values. In the 6SJ7 amp the values are sort of in the middle between trying to max out gain and going for the tamed down set up (a la Valve Wizard/merlin's values).

All this is kind of pointing me back to a cascaded V1 triode into parallel V2 into LTP then power amp...in which case my main concern goes back to the DPDT switch wiring for the parallel/cascade option. Is there a way to keep DC off of this switch, is it really a huge concern, or should I just go ahead and build it & see what happens?
User avatar
WRC34
Posts: 169
Joined: Fri Dec 19, 2014 3:40 am
Location: United States
Contact:

Re: Can't decide which way to go...

Post by WRC34 »

So I went with the original plan - V1 is a 12AX7 with a parallel/cascade switch, V2 is a hard-wired parallel 12AX7 with its own input (to completely bypass V1 for a useable clean tone), phase inverter is long tailed pair 12AX7 and outputs are cathode biased w/a shared 250R 25W resistor bypassed w/200uf/100V cap. It has a solid state rectifier, HV secondary selector switch (to choose between the 260-0-260 and 300-0-300 windings). I finished the wiring and fired it up this afternoon and it sounds excellent! If I did it again I would omit the parallel/cascade switch and hard wire for cascade since the separate parallel V2 input kind of does what I'm looking for regarding lower vs higher gain, and it would eliminate the sketchiness of the mini DPDT switch having DC voltage across it.
You do not have the required permissions to view the files attached to this post.
User avatar
WRC34
Posts: 169
Joined: Fri Dec 19, 2014 3:40 am
Location: United States
Contact:

Re: Can't decide which way to go...

Post by WRC34 »

Some readings (all numbers are DC voltage unless specified to represent resistance or wattage rating):

string of dropping resistors/choke are - UF5408 diodes > 3H 112 ohm choke > 4.7K 5W > 1.5K 5W > 10K 7W > 5.6K 2W

300-0-300 winding
B+1 384
OUTPUT
plates - 380.5
B+2 - 383.0
screens - 377.6
plate to cathode - 350.9
cathode to ground - 30.64
shared cathode resistor - 260 ohms
B+3 - 347.4 VDC
LTPI V3
plate 1/pin 6 - 211.3 (via 120K resistor)
plate 2/pin 1 - 228.1 (via 110K resistor)
B+4 - 317.7
V2 plates - 162.3 VDC (via parallel wired 110K resistor seen as 220K by 12AX7)
B+5 308.7
V1
plate 1/pin 6 - 224.0 (via 100K resistor)
plate 2/pin 1 - 224.8 (via 100K resistor)

Using the flawed but famous cathode bias formula w/my trusty pair of 6P3S-E tubes dissipation is approx 40 watts

260-0-260 winding shows 333 VDC plate voltage, 332.9 screens. Plate to cathode is 306.7 VDC and cathode to ground is 26.6 VDC which makes for approx 30 watts

Gonna try 6V6s on the lower setting and also other big guys at higher setting, most curious about EL34s

I have no schematic I just built this based on an idea but I will draw one up soon & post

Soundtrack to this build and life at the moment - https://www.youtube.com/watch?v=DX1N0j65H-0
User avatar
WRC34
Posts: 169
Joined: Fri Dec 19, 2014 3:40 am
Location: United States
Contact:

Re: Can't decide which way to go...

Post by WRC34 »

Here are some clips...

Russian 6P3S-E on higher voltage setting


Yugoslavian 6CA7 on lower setting
You do not have the required permissions to view the files attached to this post.
tubeswell
Posts: 2337
Joined: Thu Mar 27, 2008 6:42 am
Location: Wellington. NZ

Re: Can't decide which way to go...

Post by tubeswell »

WRC34 wrote: Fri Sep 22, 2017 1:35 am...6P3S-E tubes dissipation is approx 40 watts
FWIW, 6P3Se plates are rated at 20.5W (see data sheet attached) - but can probably handle 30W fairly easily (your clips sounds awesome tho')
You do not have the required permissions to view the files attached to this post.
He who dies with the most tubes... wins
User avatar
WRC34
Posts: 169
Joined: Fri Dec 19, 2014 3:40 am
Location: United States
Contact:

Re: Can't decide which way to go...

Post by WRC34 »

Tubeswell! You are a legend.

40 watts was for two tubes total, using ye olde formula of cathode -> ground voltage % actual resistance of cathode resistor (in this case a shared 260R) divide that by 2 then multiply that number by the plate -> cathode voltage = dissipation on average per tube, I then multiplied that by number of tubes (2) and voila, 40 watts. Though I have no approval this is correct, it is what I have interpreted from posts out there.

30.64 % 260 = 0.1178 % 2 = .0589 x 350.9 = 20.67 dissipation per tube for 41.35 dissipation for the pair.

And thanks! :)
User avatar
WRC34
Posts: 169
Joined: Fri Dec 19, 2014 3:40 am
Location: United States
Contact:

Re: Can't decide which way to go...

Post by WRC34 »

UPDATE - I have ventured forth, as stated, and obtained results. Regarding the use of a mini DPDT switch for cascade/parallel switching on V1 using the Zaphod Phil schematic (meant originally for 18 watt style amps), the closest I could find to an explicit warning not to do so was a post here on TAG by user tictac where he said:

"I never get this thing where hobby builders want to put high voltage DC on a mini switch. The Trainwreck guys do the same thing and there's no need to."


Now, "there's no need to" seems like an opinion rather than a cautionary warning. Having built this I would like to state publicly and definitively - DO NOT USE A MINI DPDT SWITCH FOR THIS!!! NEVER PUT B+ ON A MINI TOGGLE SWITCH!! DANGER, WILL ROBINSON, DANGER!! So there you go. It's not only unnecessary, it's INSANE.

The amp behaved ok under bench conditions but "out in the field" at a full band rehearsal the instability issues quickly appeared and rendered the amp un-useable. I have since removed the mini DPDT switch, drilled the chassis for a larger switch and replaced with a big ol' Carling DPDT switch. Everything is fine now, nice and stable.

Something interesting that happened - with the mini DPDT switch, the "Gain" pot (mounted on the back of the amp next to the switch) had no affect on the level in cascade mode. In parallel it's not meant to, but in cascade it is supposed to be a level control between the two triodes in V1. With the mini switch it behaved as if it were dimed regardless of setting. Now that the Carling switch is in place, the gain/level pot works beautifully in cascade mode to control the level between the two triodes. I changed nothing except the switch itself. Can anyone explain why this is so?

A second thing mentioned by tictac on that same post is:

"It's really easy to switch a gain stage in and out of a circuit on the AC side of the circuit.....why not do that?"

Yet, I cannot find any info on how one might go about doing this and am falling short trying to come up with a variation that might work. Perhaps you cannot truly switch between cascade/parallel on the AC side? Even so, how would you simply switch a gain stage in/out of the circuit?
tubeswell
Posts: 2337
Joined: Thu Mar 27, 2008 6:42 am
Location: Wellington. NZ

Re: Can't decide which way to go...

Post by tubeswell »

WRC34 wrote: Sun Oct 01, 2017 6:08 pm... I cannot find any info on how one might go about doing this and am falling short trying to come up with a variation that might work. Perhaps you cannot truly switch between cascade/parallel on the AC side? Even so, how would you simply switch a gain stage in/out of the circuit?
Here is one method I came up with a while back for single/parallel/cascade using a DPDT, Not prototyped yet, but it shouldn't be too difficult. Go for it
You do not have the required permissions to view the files attached to this post.
He who dies with the most tubes... wins
Post Reply