Test LL1674 transformer I/V stage for ESS DAC HAT
I made a LL1674 transformer I/V stage by this I/V DIY PCB and did some listening and comparing test on it. I have to say the sound is really good.
The style is different from most of the active I/V stages I had listened to. It's more nature. Especially in the balanced XLR output configuration.
I think the reason is that it's not only a transformer I/V, but also a passive I/V without any feedback.
I went through all different schematics. After doing a lot of listing test, I finally figure out the configuration I like the most.
With transformer I/V, though it's impossible to run DAC into real current output mode, but we can lower the I/V input impedance to to make it more close to the current mode.
I put resistors in parallel with primary winding to reduce the I/V input impedance. In this case LL1674 would be the better choice to get the finiial output level close to 2V RMS.
During my test, I/V stage was connected to PASS XP-10 solid state pre-amp. But I think, with tube amplifier, the result could be even better because of the higher input impedance.
LL1674IV1 by Ian, on Flickr
Ian
I made a LL1674 transformer I/V stage by this I/V DIY PCB and did some listening and comparing test on it. I have to say the sound is really good.
The style is different from most of the active I/V stages I had listened to. It's more nature. Especially in the balanced XLR output configuration.
I think the reason is that it's not only a transformer I/V, but also a passive I/V without any feedback.
I went through all different schematics. After doing a lot of listing test, I finally figure out the configuration I like the most.
With transformer I/V, though it's impossible to run DAC into real current output mode, but we can lower the I/V input impedance to to make it more close to the current mode.
I put resistors in parallel with primary winding to reduce the I/V input impedance. In this case LL1674 would be the better choice to get the finiial output level close to 2V RMS.
During my test, I/V stage was connected to PASS XP-10 solid state pre-amp. But I think, with tube amplifier, the result could be even better because of the higher input impedance.
LL1674IV1 by Ian, on Flickr
Ian
Last edited:
I made a LL1674 transformer I/V stage by this I/V DIY PCB and did some listening and comparing test on it. I have to say the sound is really good.
The style is different from most of the active I/V stages I had listened to. It's more nature. Especially in the balanced XLR output configuration.
I think the reason is that it's not only a transformer I/V, but also a passive I/V without any feedback.
I went through all different schematics. After doing a lot of listing test, I finally figure out the configuration I like the most.
With transformer I/V, though it's impossible to run DAC into real current output mode, but we can lower the I/V input impedance to to make it more close to the current mode.
I put resistors in parallel with primary winding to reduce the I/V input impedance. In this case LL1674 would be the better choice to get the finiial output level close to 2V RMS.
During my test, I/V stage was connected to PASS XP-10 solid state pre-amp. But I think, with tube amplifier, the result could be even better because of the higher input impedance.
LL1674IV1 by Ian, on Flickr
Ian
Hello Ian,
nice to ear you are testing the transformer version
what is the raspberry software that you are using to testing sound quality??
i´m using max2play/squeelite on my dacs. i´m always trying the others... rune, picoreplayer, volumio etc... but nothing beats max2play to my ears...
Hello Ian,
nice to ear you are testing the transformer version
what is the raspberry software that you are using to testing sound quality??
i´m using max2play/squeelite on my dacs. i´m always trying the others... rune, picoreplayer, volumio etc... but nothing beats max2play to my ears...
Hi,
I'm using Volumio for this test. I've tried picoreplayer. I know that max2play is very good. I'll try it later. Maybe I'll have some question to ask then.
Regards,
Ian
LL1674 balanced transformer I/V stage schematic
I've tried many different configurations. This is the one I like the most. Here is the schematic. Please let me know if there is any other better option.
LL1674BalancedIV by Ian, on Flickr
Ian
I've tried many different configurations. This is the one I like the most. Here is the schematic. Please let me know if there is any other better option.
LL1674BalancedIV by Ian, on Flickr
Ian
This looks very similar to quanghao's Lundahl LL1684 output boards which he first offered in 2013 for his ES9018 DAC. If you look at the schematic in the very first post here -
Super V/i converted (LL1684 )
The one significant difference (I think) is that quanghao inverts the phase from DAC to transformer primaries. As I understand it, this cancels DC offset from the +/- DAC outputs.
And he suggests that best results require the (pricey) ClarityCap ESA capacitors. But of course, Lundahl transformers are pricey too.
Yes, you are right. ESS DAC current mode is better than voltage mode. But with a transformer I/V, we can not run it at 100% current mode. Because the impedance of a transformer input can not be 0. But it should be lower as possible as can.
With the visual ground, no DC current will go through transformer windings. And theoretically no AC current will go through capacitor. That's the advantage of ESS DAC RAW balanced output. The transformer I/V is really good at this configuration.
Regards,
Ian
Last edited:
I bought this pair of LL1674 for my BIII. They are the best transformer I'm having right now.
It's 1+1:4+4, so I can lower the input impedance to make ESS DAC more to current output mode.
Please let me know if you have other recommendations.
Regards,
Ian
Have you given any thought to current steering?
Generally, as described here:
PassDiy
Or, more specifically, for example, as it pertains to the ESS9018 here:
A New Take on the Classic Pass Labs D1 with an ESS Dac
Generally, as described here:
PassDiy
Or, more specifically, for example, as it pertains to the ESS9018 here:
A New Take on the Classic Pass Labs D1 with an ESS Dac
I've also heard very good reports on the Lundahl LL1684.
I have a pair of Lundahl LL1570XL here ready to try. Ian, that proto board just fits on top of your existing DAC HAT, correct? If you have any left, I can try one pretty quickly as I have one of your ESS DAC HATS here.
In addition to the Lundahl, I can also try a pair of Onetics and Cinemags I have on hand too.
TIA!
Greg in Mississippi
I have a pair of Lundahl LL1570XL here ready to try. Ian, that proto board just fits on top of your existing DAC HAT, correct? If you have any left, I can try one pretty quickly as I have one of your ESS DAC HATS here.
In addition to the Lundahl, I can also try a pair of Onetics and Cinemags I have on hand too.
TIA!
Greg in Mississippi
Hi Greg,
Yes, that's no problem. I still have some. I'll send them you.
Regards,
Ian
I would suggest you trying my new designed ES9028Q2M DAC HAT. It a very flexible pure 9028Q2M core board with RAW balanced output. Power supplies, clock, I/V stage, controller, configuration, Sync/Async everything can be upgrade. There will be a lot of ways for effort to improve sound quality.
Actually I designed both ES9028Q2M and ES9038Q2M DAC HAT at same time. But ES9038Q2M is only capable output current as half as ES9028Q2M does. So, ES9028Q2M would be better option for a transformer I/V at this point.
Update of ES9038Q2M DAC HAT will be posted later.
ES9028Q2MDACHAT1 by Ian, on Flickr
Regards,
Ian
Attachments
Last edited:
They are for reducing the impedance of ground by connecting grounding plates together between PCB layers. We call them vias.
There are also some considerations of high frequency signal integrity, as well as EMC.
Regards,
Ian
Last edited: