Dynaco ST-70 Triode / Pentode 'conversion: A MYTH?

[Wharfcreek]

I've got another thread going on this conversion.....but it's led me to start this one. Why? Well, in that first thread I was wondering about the performance of this conversion. BUT....... in spending some time looking at the actual circuit involved in making the conversion, I'm now of the opinion that the 'triode' mode is really just still Pentode mode but in NON-Ultra-Linear. As such, I'm thinking that all the comments about being 'half-power' when in this mode are also 'perhaps' incorrect? The 'conversion' is simply removing the UL tap and connecting the screen grid (pin 4) to the Plate (pin 3) using a 100 ohm resistor. So, I'm thinking that this puts 'about' the same voltage to the screen as it might see if it was getting it from another 'tap' in the power supply. BUT, maybe some 'experts' can take a look into this and see if maybe this 'conversion' might better be titled as being 'UL/NON-UL' instead of Triode/Pentode. Just wondering.................... WC

 

[whoaru99]

...'conversion' might better be titled as being 'UL/NON-UL' instead of Triode/Pentode. Just wondering.................... WC

Some of this is a bit of sematics, don't get wrapped around that wheel. Some of it is the technical difference.

What you have normally with the ST-70 are pentodes being used in ultralinear configuration. What you propose is chaging from pentode in ultralinear operation to pentode in triode strapped operation.

As far as power output straight pentode vs ultralinear vs triode strapped, I think someone already mentioned you can see the general relatonships of that in tube data sheets.

Ultralinear is sometimes described loosely as part way between straight pentode and triode.

 

[thorpej]

For all intents and purposes, a triode-strapped pentode is a triode. Unlike a true pentode connection where the screen voltage remains relatively constant, when triode-strapped the screen sees exactly the same AC voltage fluctuations as the plate.

 

[Retrovert]

I responded in detail to this issue in your other thread here: audiokarma.org/forums/index.php?threads/dynaco-st-70-triode-vs-pentode.741035/#post-10045918

 

[Wharfcreek]

Ret's response in the 'link' in the above post is VERY informative. I clears up a LOT of the questions that surround this subject. I think I now 'get' it, and I find that relationship between the fluctuating plate voltage and the grid (when in 'strapped' mode) to be fascinating!

6, thorpe, who....... thanks for your comments as well. Clearly the simple 'Triode/Pentode' assumption here isn't really the case. But, it 'triode strapped pentode' is 'comparable' to straight triode, then that's perhaps a close call. However, the fact that the amp is really a UL Pentode in the first place, and then this simple mod takes it from UL Pentode to 'triode strapped Pentode and no longer in UL configuration'..... the simple 'pentode to triode' designation I don't think really tells the story. It appears that a straight Triode like a 2A3 or 300B would be SIGNIFICANTLY different from this 'strapped' set up......

Good info! WC

 

[thorpej]

As @BinaryMike said in his Bonseki thread:

""
I trust everyone has been disabused of the notion that triode-strapped pentodes are not ‘real’ triodes by now – their electrical characteristics are indistinguishable.
""

That's not to say that a triode strapped EL34 will behave just like a 300B, but then again there are lots of "real" triodes that don't behave like a 300B.

That said, yes, the stock configuration is UL, not pentode. If you want to get a sense of the behavioral differences, look at the plate curves for both configurations. (You'll notice that EL34 triode-strapped curves look exactly like you'd expect from a "real" triode.)

 

[Audiovet]

In support, a triode is essentially a three-element thermionic tube, with the typical characteristics of such. As each type of triode has its own figurative data because of its internal construction, they all still have typical triode characterisitcs, i.e. those of a vacuum tube with three indepedant 'contributing' elements. I consider it academic to be able to calculate what 'kind' of triode, trying to work from pentode characteristics ... I am not a physicist. I suffice with using bogey measured characteristics as I would do with any other component. Obviously the shape and size of the elements play a role, but that does not suddenly convert a triode-strapped pentode to a third kind of basic law - not from what I noticed, anyway. If you want, consider the 'strapped' version as an equivalent triode with a solid plate anode in a slightly different position (I am sure not very different). Any concept of UL or distributed load only starts appearing when there is signal between anode and screen grid. One assumes that a small strapping resistor is there simply to prevent spurious instability, that for all practical purposes the anode and screen are always at the same potential. As said, at which actual potential - (or thus far not mentioned, what effect the remaining suppressor grid has in such a case), is irrevalant in so far as one uses published data gathered from the actual application including all these effects.

 

[Pio1980]

In support, a triode is essentially a three-element thermionic tube, with the typical characteristics of such. As each type of triode has its own figurative data because of its internal construction, they all still have typical triode characterisitcs, i.e. those of a vacuum tube with three indepedant 'contributing' elements. I consider it academic to be able to calculate what 'kind' of triode, trying to work from pentode characteristics ... I am not a physicist. I suffice with using bogey measured characteristics as I would do with any other component. Obviously the shape and size of the elements play a role, but that does not suddenly convert a triode-strapped pentode to a third kind of basic law - not from what I noticed, anyway. If you want, consider the 'strapped' version as an equivalent triode with a solid plate anode in a slightly different position (I am sure not very different). Any concept of UL or distributed load only starts appearing when there is signal between anode and screen grid. One assumes that a small strapping resistor is there simply to prevent spurious instability, that for all practical purposes the anode and screen are always at the same potential. As said, at which actual potential - (or thus far not mentioned, what effect the remaining suppressor grid has in such a case), is irrevalant in so far as one uses published data gathered from the actual application including all these effects.

One difference may be in the time differential pickup of the electrons "packets" due to a finite transit time of electron particle flow. It's also possible that this can influence various simular spec'ed triodes with varied area spacing of the three elements, depending on proprietary construction.