The potential-divider connected via the 4M7 to the gate-stopper sets the DC values in the first two tubes.
It gives approx 110V on plate of the 6SL7 and also biasses up the Phase-Splitter, via the 470K from MOSFET into grid of the 6J5....
This makes the gain and phase-splitter all 'dc' coupled, so no Caps up until those coupling the grids of U3/4, the 6C33c...
The idea was to make as simple an OTL as was possible, that doesn't need constant potching with bias-adjustments, but still has good sound....
Just a thought,
If the heaters can be left to float;
Then if the top (B+) and bottom (B-) 6C33c heaters are put in series in two parallel - series chains (so I can switch one side off for reduced power).
Then the current flow through each chain (top & bottom would be the same) would this help to give equal warm up time and balance the speaker out rail (Mid point). Ok there is going to be some difference in volt drop across each tube element due to tolerance, however current flow would be the same?
What if the top was on one chain and the bottom on another the current through each chain would not be the same, could this give an unbalanced start mid point?
Regards
M. Gregg
If the heaters can be left to float;
Then if the top (B+) and bottom (B-) 6C33c heaters are put in series in two parallel - series chains (so I can switch one side off for reduced power).
Then the current flow through each chain (top & bottom would be the same) would this help to give equal warm up time and balance the speaker out rail (Mid point). Ok there is going to be some difference in volt drop across each tube element due to tolerance, however current flow would be the same?
What if the top was on one chain and the bottom on another the current through each chain would not be the same, could this give an unbalanced start mid point?
Regards
M. Gregg
Heaters in series will have the same current, no probs.
Two chains of same type valves running from same source/similar windings on mains Tx should warm through at approx the same rate--There'll always be one that will come up first or one come up a little slower than the rest, but the difference is small and happens only for a short time....
TBH, Ive done various 'Orrible' things to my bread-board over its time, like pull out one fully hot 6C33 out and shove in a stone-cold one,-
-Turn power off? Nah, Thats for wimps!!
Yup--There's a huge imbalance for a minute or two, but apart from the volume drop and a small amount of hum while the cold valve warms up there's no damage or other issues.
I'm guessing there will be DC going through the voice-coil under these extreme conditions, but its only going to be in the order of 100mA or so....
My normal DC 'Offset' with reasonably matched valves fully warmed up is about 25-30mV.
With my worst 'matched' pair Its in the order of 0.2V with the 8 ohm load connected.
Ive found its not needed to buy matched valves--Difficult with the 6C33 anyway, no-one seems to do em as I recall, anyway--They age a lot in the first few hours use so even if matched when bought, they wont be for long!.
I just put 8 ohm load on amp--resistor is fine or you can use speaker, up to you.
Put millivoltmeter accross the load and chop/change the valves round to get the lowest offset.
Sometimes, you may have a 0.1V offset, but if you switch the pair around from one socket to the other On Same Channel, then the offset may well drop to 50mV....
This is done with no signal applied BTW!
And, NO I dont 'pre-heat' the valves heaters before applying +B, --I just dont believe in it, Never had an issue with any valves caused by this treatment.
+B and Heaters at the same time, Always done it, prob. always will....
--After all, They didnt pre-heat with TV sets of the 60's and 70's and those valves lasted many thousands of hours.....
I really dont think you need worry about the difference in heating between pairs of valves, Ive seen more difference in apparent 'heat'/colour of the cathodes in different 'batch'/date valves, even though their heater-pin volts are very close/same/parallel wired....
Two chains of same type valves running from same source/similar windings on mains Tx should warm through at approx the same rate--There'll always be one that will come up first or one come up a little slower than the rest, but the difference is small and happens only for a short time....
TBH, Ive done various 'Orrible' things to my bread-board over its time, like pull out one fully hot 6C33 out and shove in a stone-cold one,-
-Turn power off? Nah, Thats for wimps!!
Yup--There's a huge imbalance for a minute or two, but apart from the volume drop and a small amount of hum while the cold valve warms up there's no damage or other issues.
I'm guessing there will be DC going through the voice-coil under these extreme conditions, but its only going to be in the order of 100mA or so....
My normal DC 'Offset' with reasonably matched valves fully warmed up is about 25-30mV.
With my worst 'matched' pair Its in the order of 0.2V with the 8 ohm load connected.
Ive found its not needed to buy matched valves--Difficult with the 6C33 anyway, no-one seems to do em as I recall, anyway--They age a lot in the first few hours use so even if matched when bought, they wont be for long!.
I just put 8 ohm load on amp--resistor is fine or you can use speaker, up to you.
Put millivoltmeter accross the load and chop/change the valves round to get the lowest offset.
Sometimes, you may have a 0.1V offset, but if you switch the pair around from one socket to the other On Same Channel, then the offset may well drop to 50mV....
This is done with no signal applied BTW!
And, NO I dont 'pre-heat' the valves heaters before applying +B, --I just dont believe in it, Never had an issue with any valves caused by this treatment.
+B and Heaters at the same time, Always done it, prob. always will....
--After all, They didnt pre-heat with TV sets of the 60's and 70's and those valves lasted many thousands of hours.....
I really dont think you need worry about the difference in heating between pairs of valves, Ive seen more difference in apparent 'heat'/colour of the cathodes in different 'batch'/date valves, even though their heater-pin volts are very close/same/parallel wired....
Hi Alastair,
How many uF do you have in each 150V + / - supply rails?
LOl..I just like to be sure I get it right.
I must admit I did not think the OTL would be as robust as you have proved!
I expected a lot of drift and offset at the speaker terminals!
It seems that there are no major considerations I'll just build.
Regards
M. Gregg
How many uF do you have in each 150V + / - supply rails?
LOl..I just like to be sure I get it right.
I must admit I did not think the OTL would be as robust as you have proved!
I expected a lot of drift and offset at the speaker terminals!
It seems that there are no major considerations I'll just build.
Regards
M. Gregg
Last edited:
I have a couple of large Sprague 3,300 200V screw-terminal jobbies I bought as a job lot of a box 20 odd NOS some 20 years ago...
One in each line, with a 1000uF 350V across the rails.
--Even after 20 years they are still good surprisingly!
As to robustness, I think this is due to the cathode-bias, I dont think a conventioal bias using a neg supply rails would be as forgiving....
One in each line, with a 1000uF 350V across the rails.
--Even after 20 years they are still good surprisingly!
As to robustness, I think this is due to the cathode-bias, I dont think a conventioal bias using a neg supply rails would be as forgiving....
I have a couple of 1000uF 450V so i guess I will use the 4700uf 250V and stick with your scheme!
Any ideas on using bypass caps across them?
Sprague are very good caps I have seen them working in electronic equipment 30yrs old and still going strong.
I think the 1000uF I have are Aerovox.
Regards
M. Gregg
Any ideas on using bypass caps across them?
Sprague are very good caps I have seen them working in electronic equipment 30yrs old and still going strong.
I think the 1000uF I have are Aerovox.
Regards
M. Gregg
Last edited:
I have some old Audio Note PIO I might use or Mundorf have some power supply film caps.
I was looking at the new socket PL7K-collet :
''GSTube.com''. Tubes, sockets etc. Tube sockets of this type.
I might get these, still collecting parts.
Did you ever think about monitoring current etc? (panel meter)
Regards
M. Gregg
I was looking at the new socket PL7K-collet :
''GSTube.com''. Tubes, sockets etc. Tube sockets of this type.
I might get these, still collecting parts.
Did you ever think about monitoring current etc? (panel meter)
Regards
M. Gregg
Last edited:
TBH, I believe there's no need in any amp for those boutique type and mega bux caps
Just good quality film/PIO caps will work well.
Ive not got current monitoring, but have checked occasionally the current. It never seems to vary by much more than 10mA....
--I think that variation is due to power-line fluctuations, as our mains here is pretty shocking!
I scoped the mains a while back, It Definately Isnt Sinus Thats for sure! Looks like what you would expect from an amp at the point of clipping. All the wave-peaks rounded off.....
Just good quality film/PIO caps will work well.
Ive not got current monitoring, but have checked occasionally the current. It never seems to vary by much more than 10mA....
--I think that variation is due to power-line fluctuations, as our mains here is pretty shocking!
I scoped the mains a while back, It Definately Isnt Sinus Thats for sure! Looks like what you would expect from an amp at the point of clipping. All the wave-peaks rounded off.....
If you are happy with 100K then by all means, use it. 100k as a gate stopper for the IRF830 in this circuit would start to compromise the performance at around 3kHz though. For a power mosfet like this i would not consider anything bigger than a few kohms. My standard choice would probably be lower than 1k, say 220 Ohms.
Regards,
Anders
Here is an article written by Allen Kimmel who pioneered this sort of MOSFET on top type gain-stage a good few years ago.
--I believe he marketed this as an upgrade to the ever-popular ST70 range of amplifiers to great acclaim....
In this article, he claims that a Gate-Stopper of up to 33K can be used quite successfully....
--Guessing the bootstrapping effect as its a Follower stage, and not a common source-stage....
IF the H.F. response was comprimised starting at 3kHz, then I'm sure I would have seen/heard this effect.....
Allen's Article-- here on web- Mu Stage Philosophy © 1993 Alan Kimmel
Here is the text part about the MOS Mu Stage--
MOS Matters
Since MOSFETs have even higher gain than pentodes, they can multiply resistor Rp's value incredibly high. Therefore, a smaller value of Rp can he used-typically, one-tenth of what Rp would he with a pentode. With a MOSFET source follower (SF), I recornmend values for Rp of 680-7.5k), depending on the current ratings of triode V1. Again, the value of Rp is not critical. The rule of thumb is higher values of Rp for low-current triodes, and low to high values of Rp for high current triodes.
Because most MOSFETs are enhancement-mode types and cannot be self biased, we must supply fixed bias to it. This does not compromise the performance in any way. We want the MOSFET's source terminal lobe at about half the B+; therefore, our fxed bias must also be half the B+. We can derive this bias voltage by using a simple voltage divider of two equal-value (say, 220k)) resistors in series. Connect one end of the voltage divider to B+ and the other end to ground with a 1-5uF capacitor across the lower resistor.
The junction of the two resistors will be the fixed bias supply point to which gate resistor Rg2 will he connected. This simple arrangement has given excellent results. However, if you do not want a voltage divider, you can use a zener string for Q1's fixed bias voltage source. Operate zeners at about 20% of their maximum current rating and use a 10uF capacitor to bypass the zener string. If you build a stereo pair or a push/pull pair of Mu Stages, the same voltage divider (or zener string) can serve both. No interaction will result.
MOSFETs are not sensitive to the value of their gate resistor. You can thus use a high value for Rg2, to minimize the load seen by V1. You will find that a value of up to 22M is okay. In such a case, reduce Cp's value to 0.02uF. Stay within 25% of Q1's maximum power rating.
--I believe he marketed this as an upgrade to the ever-popular ST70 range of amplifiers to great acclaim....
In this article, he claims that a Gate-Stopper of up to 33K can be used quite successfully....
--Guessing the bootstrapping effect as its a Follower stage, and not a common source-stage....
IF the H.F. response was comprimised starting at 3kHz, then I'm sure I would have seen/heard this effect.....
Allen's Article-- here on web- Mu Stage Philosophy © 1993 Alan Kimmel
Here is the text part about the MOS Mu Stage--
MOS Matters
Since MOSFETs have even higher gain than pentodes, they can multiply resistor Rp's value incredibly high. Therefore, a smaller value of Rp can he used-typically, one-tenth of what Rp would he with a pentode. With a MOSFET source follower (SF), I recornmend values for Rp of 680-7.5k), depending on the current ratings of triode V1. Again, the value of Rp is not critical. The rule of thumb is higher values of Rp for low-current triodes, and low to high values of Rp for high current triodes.
Because most MOSFETs are enhancement-mode types and cannot be self biased, we must supply fixed bias to it. This does not compromise the performance in any way. We want the MOSFET's source terminal lobe at about half the B+; therefore, our fxed bias must also be half the B+. We can derive this bias voltage by using a simple voltage divider of two equal-value (say, 220k)) resistors in series. Connect one end of the voltage divider to B+ and the other end to ground with a 1-5uF capacitor across the lower resistor.
The junction of the two resistors will be the fixed bias supply point to which gate resistor Rg2 will he connected. This simple arrangement has given excellent results. However, if you do not want a voltage divider, you can use a zener string for Q1's fixed bias voltage source. Operate zeners at about 20% of their maximum current rating and use a 10uF capacitor to bypass the zener string. If you build a stereo pair or a push/pull pair of Mu Stages, the same voltage divider (or zener string) can serve both. No interaction will result.
MOSFETs are not sensitive to the value of their gate resistor. You can thus use a high value for Rg2, to minimize the load seen by V1. You will find that a value of up to 22M is okay. In such a case, reduce Cp's value to 0.02uF. Stay within 25% of Q1's maximum power rating.
Last edited:
It is funny description of gyrator load. Unclear where he got that "The rule of thumb is higher values of Rp for low-current triodes, and low to high values of Rp for high current triodes." -- sounds strange to me. Also, "In such a case, reduce Cp's value to 0.02uF." sounds strange: I always thought that higher capacitance in respect to non-linear capacitances of MOSFETs is better.
Well, Rp is where the signal is produced for the MOSFET to follow...
A high current triode doesnt need such a high value resistor to develop a usable voltage signal.
Lower current triodes will need a higher Rp to get the same sort of voltage signal range....
Sounds perfectly logical to me.....
Isnt the 'gate-non-linearity' Bootstrapped anyway? Its a Source-follower not a common gain-stage...
BTW, Ive changed one of the Gate-Stoppers to 15K and left the other as 100K....
There's no apparent difference in sound from each channel....
A high current triode doesnt need such a high value resistor to develop a usable voltage signal.
Lower current triodes will need a higher Rp to get the same sort of voltage signal range....
Sounds perfectly logical to me.....
Isnt the 'gate-non-linearity' Bootstrapped anyway? Its a Source-follower not a common gain-stage...
BTW, Ive changed one of the Gate-Stoppers to 15K and left the other as 100K....
There's no apparent difference in sound from each channel....
Alastair,
Just for interest,
What effect (If any) does reducing Cp have on the sound?
Or using a film cap in that position (Leakage)?
Ouote:
MOSFETs are not sensitive to the value of their gate resistor. You can thus use a high value for Rg2, to minimize the load seen by V1. You will find that a value of up to 22M is okay. In such a case, reduce Cp's value to 0.02uF. Stay within 25% of Q1's maximum power rating
Regards
M. Gregg
Just for interest,
What effect (If any) does reducing Cp have on the sound?
Or using a film cap in that position (Leakage)?
Ouote:
MOSFETs are not sensitive to the value of their gate resistor. You can thus use a high value for Rg2, to minimize the load seen by V1. You will find that a value of up to 22M is okay. In such a case, reduce Cp's value to 0.02uF. Stay within 25% of Q1's maximum power rating
Regards
M. Gregg
If Cp is reduced in value too much, with Rs at 4.7M as in the schematic I posted, then a point will come where the LF response will be compromised....
If the cap is 'leaky' then this can affect the DC conditions of the mosfet and both the tubes too--ie, its leakage will act as a potential-divider with Rs....
'Normal' everyday film or PIO caps have sufficiently low leakage these days as not to present any problems with leakage here, there isnt a huge voltage across it anyway...
If the cap is 'leaky' then this can affect the DC conditions of the mosfet and both the tubes too--ie, its leakage will act as a potential-divider with Rs....
'Normal' everyday film or PIO caps have sufficiently low leakage these days as not to present any problems with leakage here, there isnt a huge voltage across it anyway...
Yes, Ive used a polyprop there, as well as some very old small hermetically sealed ITT PIO caps rescued from some bit of military hardware...
The main coupling-caps to the 6C33c are Russian 630V 0.47 K40y-9 hermeticaaly sealed PIO caps which are pretty cheap and excellent quality.
--You can also use conventional Polyprop/film caps here too
Best advice I can give, is Build it with Standard Parts,--Nothing mega-bux, then if you feel the need, 'upgrade' to PIO's etc....
Why waste money where there is no need?
The main coupling-caps to the 6C33c are Russian 630V 0.47 K40y-9 hermeticaaly sealed PIO caps which are pretty cheap and excellent quality.
--You can also use conventional Polyprop/film caps here too
Best advice I can give, is Build it with Standard Parts,--Nothing mega-bux, then if you feel the need, 'upgrade' to PIO's etc....
Why waste money where there is no need?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.