Good Morning Folks,
Many of you have noticed that we have introduced a new release for Trident. It is Version 3. Naturally that has generated a lot of discussion. I have been wanting to present the changes and answer the questions, so lets do it.
First where is version 2? Version 2 exists, and it is also very awesome, but it is also *very* expensive, and Brian and I decided is was probably not practical to pursue because of the very high parts costs involved. In the end we found away to get the performance gains without the degree of added expense. Version 2 had slightly a more accurate voltage reference, but Version 3 actually has lower noise. So in the end we decided to simply go with version 3.
Why the version 3 update? Version 1 works quite well, and people really enjoy it. Both of these things are true. Why touch it? But as is the case with many things, the uses for Trident necessitated a redesign. Since I needed to change Trident to support new applications (Opus and others) I took the opportunity to apply feedback to make it better.
Trident version 3 is a product of several things:
A couple of features of the new version include:
People are always asking me if it is a significant update. Well yes in many ways it is, but that does not mean I believe everyone should rush out and swap tridents for the new version. If you are happy with your current setup then you don't *need* to update it.
Owners of the previous Trident can upgrade the error amp if they have the ability, but that will not make the two versions equivalent. There are several other changes as well. If you wanted to upgrade the error amp on the Trident you have many choices. I would suggest OPA209 and OPA211 as they are both known to work quite well.
I owe a lot of thanks to a couple of skilled engineers (with far better testing gear than I have) who helped me in critically analyzing the Trident. They know who they are, and I hope they know how much I appreciate their support. I am not sure I should name them, because some of the help came from people who may wish to be anonymous. I would feel badly if I didn't at least give credit where credit is due. So thank you. I have often said that we have the best community of DIY folks a person could ask for.
So if the question is (And I get this very often) - "Russ, would you personally upgrade?" My answer is very simple. Yes I certainly would. The answer is the same for Buffalo II or Buffalo III. So now that is out of the way.
For those of you getting Trident V3 configured for Buffalo II/III notice that R4 is already installed! Do not add another one!!! Trident 3 will also be available with no R4 installed so that you can configure the CCS current yourself - but that edition is not available yet.
I hope that answers most of your questions, if you have more then please ask.
Cheers!
Russ
Many of you have noticed that we have introduced a new release for Trident. It is Version 3. Naturally that has generated a lot of discussion. I have been wanting to present the changes and answer the questions, so lets do it.
First where is version 2? Version 2 exists, and it is also very awesome, but it is also *very* expensive, and Brian and I decided is was probably not practical to pursue because of the very high parts costs involved. In the end we found away to get the performance gains without the degree of added expense. Version 2 had slightly a more accurate voltage reference, but Version 3 actually has lower noise. So in the end we decided to simply go with version 3.
Why the version 3 update? Version 1 works quite well, and people really enjoy it. Both of these things are true. Why touch it? But as is the case with many things, the uses for Trident necessitated a redesign. Since I needed to change Trident to support new applications (Opus and others) I took the opportunity to apply feedback to make it better.
Trident version 3 is a product of several things:
- We really needed a higher voltage (5V) output version for Opus
- We saw many people were forgetting to install R4 so we wanted to facilitate pre-installing R4. So now there are dual pads SMT and TH.
- We wanted a more robust Trident that can stand a higher voltage drop
- We wanted better bandwidth and better stability in high EMI environments and into very dynamic loads.
- I just couldn't leave it alone.
A couple of features of the new version include:
- A better error amp
- A better local feedback scheme at the shunt.
- A better layout with a spot for SMT CCS resistor - R4.
- Some other very small circuit tweaks(I won't give all these away).
- Much more robust transistors. They will very easily handle any TPA app.
People are always asking me if it is a significant update. Well yes in many ways it is, but that does not mean I believe everyone should rush out and swap tridents for the new version. If you are happy with your current setup then you don't *need* to update it.
Owners of the previous Trident can upgrade the error amp if they have the ability, but that will not make the two versions equivalent. There are several other changes as well. If you wanted to upgrade the error amp on the Trident you have many choices. I would suggest OPA209 and OPA211 as they are both known to work quite well.
I owe a lot of thanks to a couple of skilled engineers (with far better testing gear than I have) who helped me in critically analyzing the Trident. They know who they are, and I hope they know how much I appreciate their support. I am not sure I should name them, because some of the help came from people who may wish to be anonymous. I would feel badly if I didn't at least give credit where credit is due. So thank you. I have often said that we have the best community of DIY folks a person could ask for.
So if the question is (And I get this very often) - "Russ, would you personally upgrade?" My answer is very simple. Yes I certainly would. The answer is the same for Buffalo II or Buffalo III. So now that is out of the way.
For those of you getting Trident V3 configured for Buffalo II/III notice that R4 is already installed! Do not add another one!!!
Trident 3 will also be available with no R4 installed so that you can configure the CCS current yourself - but that edition is not available yet.I hope that answers most of your questions, if you have more then please ask.
Cheers!
Russ
Last edited:
Version 2 would have cost about twice the Version 3.
In the end the only advantage would have been a more accurate voltage reference, but version 2 would also have a bit more noise.
In the end we decided to go with version 3 - lower cost, and lower noise.
Compared to version 1 - version 3 has far more bandwidth at the shunt. It does make an appreciable difference.
Cheers!
Russ
What version of tridents I have, which comes with BIII order?
How much is the noise difference between ver 2 and ver 3?
Can I suppose that version 3 combo set for BIII will cost $40 for me to order? It would be nice if everything is changed this way. )
Last edited:
No , It is a slightly different circuit, and actually has a far better defined load.
I actually have some ideas I may try later, but nothing in the near future.
Related question... you mentioned that "Owners of the previous Trident can upgrade the error amp if they have the ability, but that will not make the two versions equivalent. There are several other changes as well. If you wanted to upgrade the error amp on the Trident you have many choices. I would suggest OPA209 and OPA211 as they are both known to work quite well."
Two questions... first, just how far down the path towards the new version does an error amp change take an original Trident?
Second, how about the OPA827 as a replacement?
TIA!
Greg in Mississippi
Two questions... first, just how far down the path towards the new version does an error amp change take an original Trident?
Second, how about the OPA827 as a replacement?
TIA!
Greg in Mississippi
1) They will still be quite different, but the Trident 1 will now have better bandwidth and be a bit more solid into extremely dynamic loads. It's very difficult to make such a comparison.
2) That unfortunately would not be a good replacement for the error amp. It really needs a split supply and the minimum supply voltage is too high.
Yes!
Replaced my 4 Tridents this morning: I use two on the B-II, and 1 for powering each (Crystek CCHD) oscillator on my async USB interface; the oscillator on the B-II is unpowered, and I am synchronous clocking from the master clocks on the USB interface.
After a few hours of playback today, there is definitely an improvement, more relaxed sound (not that it was at all noticeably nervous or strident before) and better resolution of very low level details: ambience and more complete harmonic resolution of complex instrumental timbres. Very much worth the upgrade.
Thanks Russ!
Replaced my 4 Tridents this morning: I use two on the B-II, and 1 for powering each (Crystek CCHD) oscillator on my async USB interface; the oscillator on the B-II is unpowered, and I am synchronous clocking from the master clocks on the USB interface.
After a few hours of playback today, there is definitely an improvement, more relaxed sound (not that it was at all noticeably nervous or strident before) and better resolution of very low level details: ambience and more complete harmonic resolution of complex instrumental timbres. Very much worth the upgrade.
Thanks Russ!
Which reminds me of and old suggestion which I have never proposed.
Russ, you could make the boards a fraction more DIY friendly by increasing the diameter of the leaded-holes just a bit. When you have a tight fit and a 2 layer board the removal procedure is tricky . Both suction wick and tool are struggling reaching the solder and the risk of damaging the via is imminent.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.