TU8600S with intestage

[Chul]

**** TU8600S with interstage transformer **** sorry for the typo.


Hello,

I am another fan of TU8600S and I toyed with this wonderful amp and I thought what I did may be of interest to some.

In short, I put the para-feed interstage before 300B without changing any given circuit except not using the C104 and C204.

A Fantastic fact about TU8600S is that it has these TP points that make my experiment really easy.

What I did ;

- remove C104 C204 or just cut the PCB path (in the picture)
- Use four TPs to connect the 'cap + para feed interstage transformer' (in the hand-drawn circuit..)
- I used Ludahl LL1635PP. You can use whatever 1:1 interstage. I used PP version because it has no air gap which is an advantage.
- give it some time and listen to it.
- If you don't like it you can put it back to the stock state 100%.
- If you like it move the OPTs(stock version) to the backside a little and fix the LL1635
- I used in-line connectors to compare with the original sound later.
- beware of the phase of primary and secondary, wrong connection causes oscillation interacting with cathode feedback winding of output transformer.

I was pleased to know that this mod does not change any nice appearance of TU8600S.
About the sound, well I like it but who knows about the taste of who except oneself. As a DIYer, I was once shocked by TU8600S performance and tried to improve my 2A3 SET for some different charm that can be used occasionally with TU8600S.
I can say that my TU8600S has that nice tone of 2A3 now and I am listening to more music.

Cheers,
Chul

 

[vkung]

I do not recommend LL2785B owner to do this type of modification....
The stock OPT is the bottleneck..Comments on the LL2785B design

 

[Chul]

I see and understand that the stock OPT can be the bottleneck. This implies that we are trying to get the most out of the amp, in terms of quantity which is output power.

I think TU8600S circuit is surprisingly good enough to drive 300B tubes.
Ironically we tube lovers sometimes think about this 'more power', when transistors did kill tubes for the same reason.

Tube circuits can be realized in many different ways for different tastes. There is no best.
I think TU8600S is really a nice amp with good value and I am glad that I found a way to put different tastes to this wonderful amp.

Enjoy your TU8600S folks.

 

[Chul]

I do not recommend LL2785B owner to do this type of modification....
The stock OPT is the bottleneck..Comments on the LL2785B design

This mod can be tried with stock or Lundahl both. I hope I can do it soon.

 

[vkung]

I just ask you to try LL2785B before the modification.

 

[Chul]

I just ask you to try LL2785B before the modification.

I see. I got the quote from you. I am just having enough time with the stock OPT. You know this TU8600S gives me such an impression that almost discourages my DIY hobby.

 

[vkung]

it is a huge improvement..

 

[paczeltf]

In short, I put the para-feed interstage before 300B without changing any given circuit except not using the C104 and C204.(...)
I used Ludahl LL1635PP. You can use whatever 1:1 interstage. I used PP version because it has no air gap which is an advantage.

I don't have the schematics for the TU-8600 but I have a few generic comments.

- The additional interstage transformer will almost certainly change the gain of the driver stage and it may not be able to provide enough undistorted signal for the power tubes.

- It is also possible that the positive voltage swing created by the transformer will cause damage to the driver tubes - 12AU7 is a small signal tube.

- The change to the gain of the driver stage along with the phase shift introduced by the interstage transformer will affect the global negative feedback, which requires changes in amount as well as phase compensation.

- A no-air-gap parafeed interstage transformed may not pair well with an air-gapped SET output transformer. Mixing different design approaches tends to highlight the cons of the different designs rather than their advantages.

The bottom line is that a proper implementation of an interstage transformer would require a significant redesign of the amp.

 

[Chul]

I don't have the schematics for the TU-8600 but I have a few generic comments.

- The additional interstage transformer will almost certainly change the gain of the driver stage and it may not be able to provide enough undistorted signal for the power tubes.

- It is also possible that the positive voltage swing created by the transformer will cause damage to the driver tubes - 12AU7 is a small signal tube.

- The change to the gain of the driver stage along with the phase shift introduced by the interstage transformer will affect the global negative feedback, which requires changes in amount as well as phase compensation.

- A no-air-gap parafeed interstage transformed may not pair well with an air-gapped SET output transformer. Mixing different design approaches tends to highlight the cons of the different designs rather than their advantages.

The bottom line is that a proper implementation of an interstage transformer would require a significant redesign of the amp.

Let me honestly comment from what I experienced.

- The gain of this amp is just more than enough. I felt almost no change.

- I toyed with many parafeed from pure experimental fun, including autoformer, Magenquest cobalt, and Sowters and bifilars. Even small toroidal power trsnaformers fro parafeed output. I respect the designer and his work very much. But no designer can explorer all the possibilities, unless on simluations. So I invited my 'better eared' friend of mine and he liked it.

- The bottome line is what you hear over long time and how you like what you hear.

You are impressive in that you don't seem to have experienced this wonderful amplifer and you are saying from your 'may' approach. I thought my DIY is useless after hearing my TU-8600S. That respect was in my mind when I was tempering with the amp.

Elekit amp is a kit and it may look like a DIY but actually a product tightly designed and packed. Nothing like the DIYs on this forum. It is so well designed mechanically and electronically. I am just glad about myself that I found some spot to enjoy this amp 'my way'. You can enjoy your way.

For the interstage used in my mod, the best was bifilar ones.

 

[paczeltf]

- The gain of this amp is just more than enough. I felt almost no change.

You seem to have misunderstood that I wrote. The overall gain of an amp is a personal preference to a big extent, but a gain in a particular stage in that amp is not.

You are impressive in that you don't seem to have experienced this wonderful amplifer and you are saying from your 'may' approach.

You do not have any technical arguments and you are getting personal. The points that I raised apply to all changes from an R/C coupled design to an interstage transformer, especially if global feedback is involved.

If you follow the trial-end-error route, no problem, it is your amp in the end. But let me raise a warning sign for those who may be considering similar changes.

 

[Chul]

You seem to have misunderstood that I wrote. The overall gain of an amp is a personal preference to a big extent, but a gain in a particular stage in that amp is not.



You do not have any technical arguments and you are getting personal. The points that I raised apply to all changes from an R/C coupled design to an interstage transformer, especially if global feedback is involved.

If you follow the trial-end-error route, no problem, it is your amp in the end. But let me raise a warning sign for those who may be considering similar changes.

Audio is all about a personal preference because it is not about instrumental precision. But the gain is not a personal preference but a simple measurement.
That is why there is a volume control for loudness preference.

In fact, the effect of using interstage is normally toward 'increased feel' due to the dynamic character of the transformer.
And the insertion loss of the interstage transformer is negligent due to this effect.

A warning sign for DIYer...warning for what?

Please go ahead build and listen to this amp then we can talk.
Or you can measure the effect of the interstage transformer with a global feedback and let us know.

Do you know that the most famous output transformers usually show the worst waveform on the oscilloscope?
Maybe you can find out what it is like with TU-8600S.

 

[paczeltf]

Please go ahead build and listen to this amp then we can talk.

I am not going to build an amp that lacks circuit design. I know this attitude well from various FB groups and forums and got tired of reading them through long ago. This was my last post to this thread. I am sorry but I don't have time and motivation for this.

P.S. This forum has been reasonably free of trolls and I hope it remains like that.

 

[Steve Morley]

Just a thought guys:
A Tubelab SE 300B amp (forum on this site) might be a better place to start for some serious experimenting. A board is about 30 bucks and you can add transformers of your choosing and hack the board if desired. It would save carving up the chassis and board of a TU8600S. At least until you have a working design.

BTW, can't speak to the difference a Lundahl OPT makes on the 8600 but it made a HUGE difference on the TU8340VK I built. But note; the Lundie sounded awful when first installed. At 50 hours it sounded better than the stock OPT and by 100 hours it blew the stock trannie away in both UL and Triode mode.

Cheers, Steve