Mikaels KT88SE & VTV Octal Linestage
Hello all at diyaudio,
My first post so please excuse me should you think my thread a little stupid
Just waiting for some bespoke transformers from Sowter, then I can finish Mikaels KT88SE amp; to which I have a couple of queries if anyone can help. (?)
1) The amp schematic includes a pot. I would like to remove this and have a power amp only (To go with my VTV linestage). Would this be OK? and if so would I need to omit the pot and add in a 100k resister across the input?
2) Regarding the Linestage (Issue 11 of VTV), the article states that the B+ is 370v. The same editions goes further in testing a number of 6sn7's (GT's, GTA's, GTB's, CV181's and VT231's) using this linestage. I understand that the GT's, CV181's and VT231's have a lower max plate voltage (around 250v - 330v). The reason I ask is that at present I am using EH gold pins but would like to use a couple of Ken-Rad VT231's that I have, but i'm a little worried about catastrophic explosions or arching.
Strictly hobbiest only i'm afraid, self educated. But I have built a WAD KEL84, a number of speakers and also Upgraded Quad 303's (I like) and 405's (Not so much).
Would welcome some comments and opinions on the VTV & KT88, any good?
Thanking you all in anticipation.
Regards
Adam
Hello all at diyaudio,
My first post so please excuse me should you think my thread a little stupid
Just waiting for some bespoke transformers from Sowter, then I can finish Mikaels KT88SE amp; to which I have a couple of queries if anyone can help. (?)
1) The amp schematic includes a pot. I would like to remove this and have a power amp only (To go with my VTV linestage). Would this be OK? and if so would I need to omit the pot and add in a 100k resister across the input?
2) Regarding the Linestage (Issue 11 of VTV), the article states that the B+ is 370v. The same editions goes further in testing a number of 6sn7's (GT's, GTA's, GTB's, CV181's and VT231's) using this linestage. I understand that the GT's, CV181's and VT231's have a lower max plate voltage (around 250v - 330v). The reason I ask is that at present I am using EH gold pins but would like to use a couple of Ken-Rad VT231's that I have, but i'm a little worried about catastrophic explosions or arching.
Strictly hobbiest only i'm afraid, self educated. But I have built a WAD KEL84, a number of speakers and also Upgraded Quad 303's (I like) and 405's (Not so much).
Would welcome some comments and opinions on the VTV & KT88, any good?
Thanking you all in anticipation.
Regards
Adam
100K resistor across the input should work in lieu of the 100K pot. Mikael's SE KT88 is one of the best amp I have ever built.
I built a similar linestage to the VTV octal (is this the grounded cathode into a cathode follower linestage). Am using 12sn7 and the plate voltage is around 230V. Don't forget to lift the ground by about 50V.
HTH.
I built a similar linestage to the VTV octal (is this the grounded cathode into a cathode follower linestage). Am using 12sn7 and the plate voltage is around 230V. Don't forget to lift the ground by about 50V.
HTH.
Hi Alex,
thanks for the reply.
VTV linestage is as:-
http://www.geocities.com/northbirdten/line_stage.htm
Sorry, what do you mean when you say dont forget to lift the ground by 50v ?
thanks for the reply.
VTV linestage is as:-
http://www.geocities.com/northbirdten/line_stage.htm
Sorry, what do you mean when you say dont forget to lift the ground by 50v ?
By the way, if you want to lower the B+ a bit, you can change the 5AR4 rectifier for something like a 5U4, provided your rectifier heater winding (the 5v one) can handle a 3 amp draw. That will knock things down 20v or so. I don't think the 6SN7's your using, however, should not have a problem.
6SN7
The maximum voltage ratings for 6SN7 means the dc voltage from anode to cathode during steady operation, so don't worry about 370V at startup. Sometimes you see ratings for 1960s/1970s valves for the maximum supply (Vab) or maximum Cold Anode Voltage Va0 (eg Telefunken specs) or Max Voltage at Cutoff (Russian specs). These all mean the same: max volts applied when no anode current flows. If the 6SN7 had such a rating it would be about 500V.
Raising the heater voltage to 50VDC is a hum countermeasure, as noted. The 6SN7 has has a long production history, and some variants will need this, while really good ones (Like GEC, Sylvania GE, RCA) will not (Can run at heater grounded one side or 100 Ohm resistors from each side to ground). Some later Eastern kind will still hum even with raised heater voltage - because one of the grids is internally wired too close to the heater pin next door to it. If this happens, you'll need to run the heaters dc (or get an expensive 6SN7)
I see the VTV linestage has a cathode follower output. Many folks here won't agree, but I believe cathode followers are a waste of a valve, and sound poor. If you are given to experimentation, try replacing the follower with a 600V rated MOSFET source follower or a 450V BJT. Try comparing the sound.
The maximum voltage ratings for 6SN7 means the dc voltage from anode to cathode during steady operation, so don't worry about 370V at startup. Sometimes you see ratings for 1960s/1970s valves for the maximum supply (Vab) or maximum Cold Anode Voltage Va0 (eg Telefunken specs) or Max Voltage at Cutoff (Russian specs). These all mean the same: max volts applied when no anode current flows. If the 6SN7 had such a rating it would be about 500V.
Raising the heater voltage to 50VDC is a hum countermeasure, as noted. The 6SN7 has has a long production history, and some variants will need this, while really good ones (Like GEC, Sylvania GE, RCA) will not (Can run at heater grounded one side or 100 Ohm resistors from each side to ground). Some later Eastern kind will still hum even with raised heater voltage - because one of the grids is internally wired too close to the heater pin next door to it. If this happens, you'll need to run the heaters dc (or get an expensive 6SN7)
I see the VTV linestage has a cathode follower output. Many folks here won't agree, but I believe cathode followers are a waste of a valve, and sound poor. If you are given to experimentation, try replacing the follower with a 600V rated MOSFET source follower or a 450V BJT. Try comparing the sound.
Thanks for the replies gents,
In honesty, its all a bit beyond me though.
Rod, I really like the sound of mine as it is, certainly wouldn't describe it as poor. But then again I probably haven't been exposed to really good circuits as you obviouly have.
Would you be prepared to expand on your follower replacement?
Yes, I am given to experimentation but in all honesty is usually on a 'painting by numbers' basis if you know what I mean.
Still confused with these 6sn7's though. Almost tempted just to put the Ken-rads in and see what happens. Surely, if VTV tested them in the linestage they should be OK?
In honesty, its all a bit beyond me though.
Rod, I really like the sound of mine as it is, certainly wouldn't describe it as poor. But then again I probably haven't been exposed to really good circuits as you obviouly have.
Would you be prepared to expand on your follower replacement?
Yes, I am given to experimentation but in all honesty is usually on a 'painting by numbers' basis if you know what I mean.
Still confused with these 6sn7's though. Almost tempted just to put the Ken-rads in and see what happens. Surely, if VTV tested them in the linestage they should be OK?
The source follower replacement for a cathode follower is neatly explained in the MOSFET follies page:
http://www.geofex.com/Article_Folders/mosfet_folly/mosfetfolly.htm
Don't worry about using the 6SN7 in that circuit! It will be quite safe.
http://www.geofex.com/Article_Folders/mosfet_folly/mosfetfolly.htm
Don't worry about using the 6SN7 in that circuit! It will be quite safe.
Hi Rod,
2 things that I have done with my line-stage.
1) Replaced by EH gold pin 6SN7's with the Ken-Rad VT-231's. OMG, I cannot believe the difference, I was aware of the benefit's of tube rolling but this was a big suprise. Completley different, most notably balance. The bass has increased substantially and in general sounds like a completley different amp.
2) Replaced the Cathode Follower with the Mosfet as per instructions within MOSFET Follies. Not too sure about this one Rod. Yes it's different, with a little more detail and perhaps more 'modern' sounding. But the fullness had gone and it sounded a little clinical. Do these comments sound about right?
I've changed back to using the Tube as Cathode Follower for the time being! (You must think me stupid!)
Maybe I haven't given the MOSFET time to 'run-in' but i'm happy with the amp as it is.
Surely, this is diyAudio in it's purist sense. Perhaps technically not correct, but it suits me just fine.....
Regards
Adam
2 things that I have done with my line-stage.
1) Replaced by EH gold pin 6SN7's with the Ken-Rad VT-231's. OMG, I cannot believe the difference, I was aware of the benefit's of tube rolling but this was a big suprise. Completley different, most notably balance. The bass has increased substantially and in general sounds like a completley different amp.
2) Replaced the Cathode Follower with the Mosfet as per instructions within MOSFET Follies. Not too sure about this one Rod. Yes it's different, with a little more detail and perhaps more 'modern' sounding. But the fullness had gone and it sounded a little clinical. Do these comments sound about right?
I've changed back to using the Tube as Cathode Follower for the time being!
(You must think me stupid!)Maybe I haven't given the MOSFET time to 'run-in' but i'm happy with the amp as it is.
Surely, this is diyAudio in it's purist sense. Perhaps technically not correct, but it suits me just fine.....
Regards
Adam
Have fun rolling!
If you're using a Valve power amp AND your line level cables are reasonably short (<2m) you could try running a CD Player and/or RIAA stage stright into it without the line amp. Use 20 to 50K volume control (make a stepped one yourself using LORLIN rotary switches from Maplin or Farnell) and use 1206 chip resistors (lowest inductance, no magnetic materials).
some precautions to observe with source followers:
- minimize the wiring, especially the gate (<60mm).
- use a stopper in the gate (chip resistor 1206 or 0805 270 to 1K ohm)
- high frequency film capacitor 10n..470n right at the drain, with short wire to ground (some blank FR4 needed really to form a plane)
- wire the load resistor and driving triode's cathode resitor to the same ground area (FR4 PCB plane)
MOSFETs have extended high frequency performance, and can oscillate in the MHz region if untamed
If you're using a Valve power amp AND your line level cables are reasonably short (<2m) you could try running a CD Player and/or RIAA stage stright into it without the line amp. Use 20 to 50K volume control (make a stepped one yourself using LORLIN rotary switches from Maplin or Farnell) and use 1206 chip resistors (lowest inductance, no magnetic materials).
some precautions to observe with source followers:
- minimize the wiring, especially the gate (<60mm).
- use a stopper in the gate (chip resistor 1206 or 0805 270 to 1K ohm)
- high frequency film capacitor 10n..470n right at the drain, with short wire to ground (some blank FR4 needed really to form a plane)
- wire the load resistor and driving triode's cathode resitor to the same ground area (FR4 PCB plane)
MOSFETs have extended high frequency performance, and can oscillate in the MHz region if untamed
I imagine that cathode followers are worse than ordinary transistor or MOS followers is related to the low value of gm (ie: slope S, mutual conductance).
With low gm, the distortion increases with load (including cable capacitance, for instance)
Vout/Vin = RL /(RL + (1/gm))
Since gm varies with the instantaneous current - ie it's nonlinear - (in valve, BJT or MOSFET), the larger the value, the lower the distortion term in the output.
For a BJT, the gm is 1/re, or
Ie/(KT/q)
which amounts to 40mA/V per mA of emitter current (ie gm = 400mA/V at Ic = 10mA). It's less than that for a MOSFET, but still much greater than the 1 to 25mA/V range available in normal valves, or 50mA/V in the E810F which is at or near the highest possible.
In very badly implemented cathode followers (eg an ECC83 with 1mA anode current driving 4m of cable followed by a 10K load) the cathode forms a significant potential divider (~10:1 at LF) with serious nonlinear components and useless sound quality.
All this applies to a cascode, too; I prefer cascodes with ordinary bipolar transistors "at the top"- soundwise to anything else.
With low gm, the distortion increases with load (including cable capacitance, for instance)
Vout/Vin = RL /(RL + (1/gm))
Since gm varies with the instantaneous current - ie it's nonlinear - (in valve, BJT or MOSFET), the larger the value, the lower the distortion term in the output.
For a BJT, the gm is 1/re, or
Ie/(KT/q)
which amounts to 40mA/V per mA of emitter current (ie gm = 400mA/V at Ic = 10mA). It's less than that for a MOSFET, but still much greater than the 1 to 25mA/V range available in normal valves, or 50mA/V in the E810F which is at or near the highest possible.
In very badly implemented cathode followers (eg an ECC83 with 1mA anode current driving 4m of cable followed by a 10K load) the cathode forms a significant potential divider (~10:1 at LF) with serious nonlinear components and useless sound quality.
All this applies to a cascode, too; I prefer cascodes with ordinary bipolar transistors "at the top"- soundwise to anything else.
Thanks for your time Gents,
Funny about Cathode Followers sounding horrible.
I wasn't aware until I completed the linestage, then read loads of stuff saying that they wasn't worthwhile, although it was a classic circuit.
Thought i'd wasted my time and money until I heard it though, in particular with the Ken-Rad's. No doubt this is one of the benfits of DIY audio - do what you want as long as you like the sound.
I'm going to save for GEC KT88's though, after the difference the vt231's made I'd be a fool not too. Trouble is we are looking at £300-£400 for a pair
Regards
Adam
Funny about Cathode Followers sounding horrible.
I wasn't aware until I completed the linestage, then read loads of stuff saying that they wasn't worthwhile, although it was a classic circuit.
Thought i'd wasted my time and money until I heard it though, in particular with the Ken-Rad's. No doubt this is one of the benfits of DIY audio - do what you want as long as you like the sound.
I'm going to save for GEC KT88's though, after the difference the vt231's made I'd be a fool not too. Trouble is we are looking at £300-£400 for a pair
Regards
Adam
Actually Rod,
Having just re-read your post, I note that you state to insert a stopper resister and a cap also in addition to the MOSFET.
I didn't do this, I just followed the instructions on MOSFET Follies, i.e. FET replaces triode no other additional items.
Maybe this is why it didn't sound so good?
Having just re-read your post, I note that you state to insert a stopper resister and a cap also in addition to the MOSFET.
I didn't do this, I just followed the instructions on MOSFET Follies, i.e. FET replaces triode no other additional items.
Maybe this is why it didn't sound so good?
I should have mentioned it at the time. MOSFETs with oscillations (which often only occur during part of the waveform) sound horrible.
It's not a certainty that it was happening with yours, but I thought I should note it in case you're in the mood to fiddle with the circuit!
alternatively, the cathode follower may blend in better with the other parts of your sound chain - hard to say without a lot of trial & error - but some folks are restless & like to cut & try!
If you do build the source follower again, try a variety of 1206 chip resistors (from 220Ohm to 1K or more) in the gate stopper position. Or use carbon composition (new components). Don't use ordinary resistors under any circumstance - these are inductive & will provoke more oscillation than they stop (or rather, they have the potential to do that)
Mount the resistor 10mm from the MOSFET gate lead, on a tiny bit of plain copper FR4 board scored across.
Beware of success in one piece of tuberolling luring you into spending big sums on rare valves - can go the wrong way on you. Try the Russian "Genelex" KT88, which a number of commentators speak of in terms comparable to the GEC.
It's not a certainty that it was happening with yours, but I thought I should note it in case you're in the mood to fiddle with the circuit!
alternatively, the cathode follower may blend in better with the other parts of your sound chain - hard to say without a lot of trial & error - but some folks are restless & like to cut & try!
If you do build the source follower again, try a variety of 1206 chip resistors (from 220Ohm to 1K or more) in the gate stopper position. Or use carbon composition (new components). Don't use ordinary resistors under any circumstance - these are inductive & will provoke more oscillation than they stop (or rather, they have the potential to do that)
Mount the resistor 10mm from the MOSFET gate lead, on a tiny bit of plain copper FR4 board scored across.
Beware of success in one piece of tuberolling luring you into spending big sums on rare valves - can go the wrong way on you. Try the Russian "Genelex" KT88, which a number of commentators speak of in terms comparable to the GEC.
Re: Re: 6SN7
Audiophoolz dispensing the usual audiophoolish "wisdom". That's what happens when these yutzes get a little bit of knowledge. They heard that cathode followers have 100% negative feedback and they go: "OMGzorz! Negative feedback!" They believe that all negative feedback is the Spawn of Satan.
Either that, or they read all about how cathode followers have Lo-Z output impedance and then think they can drive insane lengths of highly capacitive cable with one and are disappointed in the results. What they fail to understand is the difference between output impedance and load impedance. The cathode follower is a good Lo-Z source into a Hi-Z load. It doesn't work so swell when connected to a Lo-Z load.
Followers are the least offensive stage, allow direct coupling to finals that eliminates the blocking problem, and can source enough current to prevent slewing at the upper end and improve transient overdrive behaviour. They sound just fine and are well worth that extra hole. Need more drive for going Class A*2? Use a source follower (OMG! Sand!). That'll give the audiophoolz a real case of apoplexy.
alexg said:
Audiophoolz dispensing the usual audiophoolish "wisdom". That's what happens when these yutzes get a little bit of knowledge. They heard that cathode followers have 100% negative feedback and they go: "OMGzorz! Negative feedback!" They believe that all negative feedback is the Spawn of Satan.
Either that, or they read all about how cathode followers have Lo-Z output impedance and then think they can drive insane lengths of highly capacitive cable with one and are disappointed in the results. What they fail to understand is the difference between output impedance and load impedance. The cathode follower is a good Lo-Z source into a Hi-Z load. It doesn't work so swell when connected to a Lo-Z load.
Followers are the least offensive stage, allow direct coupling to finals that eliminates the blocking problem, and can source enough current to prevent slewing at the upper end and improve transient overdrive behaviour. They sound just fine and are well worth that extra hole. Need more drive for going Class A*2? Use a source follower (OMG! Sand!). That'll give the audiophoolz a real case of apoplexy.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.