Good hearing from you Todd. Your mods look quite good. JFET's have been popular for high end audio for quite some time. I used them in a tube preamp I built many years ago.
Modding the PS Audio Directstream DAC MK1 and MK2
I would like to continue discussing all the mods we have created for these two DAC's as it seems PS Audio is no longer allowing discussions about them on their forum.
For reference here is a link to the closed discussion.
https://forum.psaudio.com/t/modding-the-directstream-dac-mki/2680
Post removed |
Some other news, this summer I was able to get Tom C of Neurochrome Audio to do a special version of his Modulus "error correcting" amp boards (thanks Tom!). Notably, we eliminated the OA input buffer, and the input coupling cap (and feedback cap!) and added back a DC servo since its now DC coupled. IMO this should allow the most transparent Modulus you can do basically. I’ve been listienng to it for months (its playing here now with a new proto preamp) and it’s quite good in fact. Note that, with the OA input buffer gone the boards are single ended input only, which is fine. The amp modules themselves are also single ended anyway, as most amps are internally. And again, its fine. However, for best results it does need the best OA we can manage for the "error correcting" portion, and for the DC servo. Most industrial OA are evil in comparison; discrete ones generally less so at least. I socketed my build and prefer the Sparkos SS3602 in the error correcting location and a Muses01 in the DC servo. Muses OA appear standard old industrial but they are far from it, employing a Cu leadframe for example. They put a lot of effort into these in fact and you can definitely hear it. The 01 is a JFET input and is very clean and quiet, also good for a DC servo IMO. The 02 is a bipolar input and "rocks" a bit more in comparison IME. But not as good for a DC servo due to input bias currents. I have mine in a Naim NAP180 chassis with what I call a "super" PSU in it. I think I would recommend a good SMPS for it if DIY. As Tom says, with this much error correcting horsepower available the boards are somewhat immune to PSU, but still, better PSU is better in general. If you want to DIY your own, ask Tom for the "Todd King" edition Modulus boards. I can recommend them. TK |
Hello folks, still here occasionally :) Thought I'd share some things. The Vocm mod with JK high nickel trafos is quite good and would likely satisfy the majority of listeners IMO. I will say though, the pinnacle of DS DAC mk1 sonic performance is, Diff amps topology mod (also found some diff amps I prefer to the AD803x), along with a bi-polar PSU that then allows for no caps or trafos on the output. This also allows Vocm to be=ground, so no mod needed for it anymore. It's not trivial to do but can be done, I'm listening to mine here now with it. Now we're really talking, now there is no audible distortion (quite rare IME) and you are hearing the fw performance basically. Any further improvement would be upgraded fw. I asked if we could get a Mk2 version improved fw for Mk1 and the answer was no. BTW I prefer the older Sunlight fw on mine; it's more open and dynamic and alive sounding. It's playing here now. I recall someone on the forum had done some topology mod effort, but they claimed they didn't like it. I don't know what the problem was there, and maybe I had also misguided somewhat with some poorly suggested values at the time. But whatever, I don't have the time or energy these days to try and maintain and acceptable S/N ratio of a web forum and that's when I called it quits there basically. The truth of it is what I posted above. TK |
Agree. Thanks for posting this. We feel these transformers are a much better investment than say $1000 power cables and it’s immediately noticeable and measurable. The MK1 upgrade makes a lot more sense for many instead of the large investment in the MK2 which they still have not moved past Beta testing even thou they are and have advertised it as a "ready for market" product however we feel the MK2 transformers are a must over the currently configured Edcor’s being fitted in the MK2. Look at the sweeps we have done of each. |
The internet can be a weird place. I've generally seen negative comments about PS Audio DACs outside of the PS Audio forum so it's a little surprising that you've tripled sales after moving away from the forum. But it is possible that the website helped people feel like the transformers are more of a business and not just a thing where you message some guy and hope to get something back after dropping $600+. |
That may be true however without the Vocm mod and our nickel transformers in the MK1, it’s not a fair comparison. We have both the fully modified MK1 and the MK2 with our transformers, the input grounds fixed and 3M RF shielding setup to do instant blind A/B comparisons and most who listen here cannot pick a winner. They are that close. I can pick the MK2 most of the time however I know exactly what to listen for not necessarily the best sounding. They are both great sounding. Hands down the build quality of the MK1 is far better than the MK2. We are pretty disappointed in the build quality of the MK2 since we have been modding many of them here. Every one of them have / had build issues. |
Below is a Post by Elk on the PS Audio Forum. He can’t even be honest with the forum members over there. He knows I can’t post on the PS Audio Forum to correct false information. He did in fact delete any post that praised our mods and / or post that showed any flaws in their product and how to fix them since the new MK2 was launched. As a matter of fact, PS Audio is currently using my fix for the latest FLAW the dreaded input grounding clips.
The last one was the one I posted here on this Forum. https://forum.audiogon.com/posts/2586529 I have many screenshots to prove all of this. He even locked down the modding topic after this was posted. I hope he will come back one day. The APS Mod for MK2 he did was a great improvement for me." "None of his posts were removed from this forum other than those which contained personal attracts on others. Most of his deleted posts were those he personally deleted when he was mad at someone, often one of his customers. I publicly requested he restore them as they were of interest to others. He refused. We did move his product posts to his dedicated sales thread. With this clarification we will not be discussing him or any other member of the forum, past or present. Please feel free to start another thread to discuss mods if interested in doing so."
|
Thanks for the response. I’m not sure what "I’m doing wrong" either. I’m not tech illiterate, I promise. So you don’t see my message in your inbox? FWIW I don't see where to check inbox in the forums. You have to go back to main audiogon website (delete "forum" in the url) to view messages When you say backed up till November, does that mean I need to reserve a spot now or just wait and check back in November?
|
Not sure what you’re doing wrong as no one else is having issues connecting with me using ether method you state. Have communicated with many people with no issues. I’ll PM you now. We are backed up till November. We did add the Ag 1948. A little bright for our taste however some may like them. We also added a network isolator for streaming. We are quite surprised how much improvement it makes.
|
I’m looking to get these upgrades for my MK1 and have tried getting in touch here via private message and the contact form on the website. No luck yet. I see today I have another transformer option. I thought I was set on the nickel but see the is now the LUNDAHL 1948Ag Pure Silver windings Transformers. How do they compare between the two? |
I don’t recall us making them exact statements. The Custom nickel transformers do in fact increase the output level by 6dB because they are and were designed as a 1:2 transformer instead of the original 1:1 transformer that are installed in the MK1. We did state that The Custom nickel transformers have a much lower output impedance than the MK2 Edcor XS4400 transformers which places the secondary windings in a series configuration in order to achieve a 1:2 output which creates a very high output impedance. This was kind of a cheat as the Edcor XS4400 was not designed to do so. The Custom nickel transformers are designed from the ground up specifically for each model, The MK1 or the Mk2. The lower impedance is a plus as it in theory should have a lower noise floor going directly into a power amp. We never tested this as we recommend going through a pre-amp. |
Hi Jeff, one of the reasons I'm considering your transformer upgrade for the MKI has to do with connecting the DAC directly to my Pass amp. I seem to recall you saying that your transformer upgrade increases the output voltage, which helps when you connect the DAC directly to a power amp. I haven't noticed anyone talking about this lately so can you please comment on this? |
Unbelievable... It took PS Audio forum members over a week to discover and discuss this problem after my post exposing this issue was deleted by Elk. The ones who have the tabs missing or loose... This is not an easy fix as they rip off the copper foil when they come loose. All units I have worked on here had this issue. Even brand-new ones that have just shipped. I knew it would eventually catch up to them and be exposed and many questions asked by unhappy customers. PS Audio should have done the right thing from the beginning instead of trying to hide the issue. https://forum.psaudio.com/t/directstream-mk-2-observations/31566/2602
|
Another funny post from PS Audio forum: Also the thing with measurement, If I have a predetermined bias before hand, I can make the measurement results to support the said bias. I’d take any audio measurements with a huge grain of salt unless the methology is published and can be verified by an unbiased party. What??? How ridicules is that? |
You can read all about it here: https://forum.psaudio.com/t/directstream-mk-2-observations/31566/1936 and https://forum.psaudio.com/t/directstream-mk-2-observations/31566/2468
|
Posted on other site. Not sure why someone would want to bad mouth PSA so much except for profit or just to be mean. Life is too short for that kind of shtuff. Also without PSA his transformer business would be non existent so need to be careful before shooting oneself in the foot.
What makes him think I’m badmouthing PS Audio??? I was asked by Elk, the moderator on that site to post these measurements. That’s all I did was post actual unbiased measurements of both the MK1 and MK2. |
Not to trivialize noise measurements, but this sort of does show how there is not a 1:1 of noise and Sound quality. Now that being said there are few reports of some amps this causing issues with, but that is a known factor and reported in the Stereophile mag. I for one love the SQ in my setup, could care less on the measurements. I have a scheduled appt. to put in my APS transformers from JK next week. Going to let burn in while away for a few days I am out of the house. Hope it does not bother the rest at home. But I should get 250+ hours burn in with them. |
Here is a FFT plot of the MK2 with the new Massive firmware. Note the distortion is worse than with the last firmware. This new massive firmware is almost impossible to measure the actual noise floor. MK2 with Massive FW:
The MK1 has better measurements. Note the better dynamic range and lower noise floor of the MK1. Note this one has a bad transformer on the right channel. |
Over the past month we have been upgrading many of the PS Audio Directstream MK2 DAC with our nickel transformers and have discovered very poor build quality on every one. We ended up having to offer to fix these defects as we don’t like sending units back half baked. We re-route the WiFi antenna Cable, Repair the input chassis grounding and fix any hardware issues (stripped wrong screws etc). We even found PC board standoffs 1 turn out loose :(
This defect shows that the i2s inputs chassis grounding spring is stuck behind the plastic insulator making it inoperable. The inputs end up with no chassis ground. Top view
This shows the mis-aligned paint mask keeping all inputs from being chassis grounded. Note the grounding springs are insulated by the paint. The only way to fix this is to scrape the paint away.
This show the RF shielding fingers not install properly. Most all units had most of them missing or none at all.
This shows how we fix the i2s grounding issue. Note now the chassis grounding spring can make contact with the chassis.
We remove the insulator and cut the right side down so the grounding spring will clear. Note the cut line.
After the cut |