Jump to content
Sign in to follow this  
Ben-M

Group Buy for full ATX Linear PSU from Teradak

Rate this topic

Recommended Posts

Thanks for the update Ben, rather they be 100% sure they have ironed out [any possible] gremlin than ship the PSU sooner.

Share this post


Link to post
Share on other sites

Delayed Communication:

Hey guys, as many of you know I now live in China, but right now my wedding is going on and all of my family from Canada is here in China. Chinese weddings are overwhelming and over the top, so I haven't been in regular communication with anyone for about 1 week. I will effectively be occupied and away until Feb. 9th, so during this time I won;t be answering emails or PMs. I thank all of you in your understanding about this and will take care of everything when I return.

I do know that the new parts for the possible solution have been mailed to the buyer and will be installed soon. After that we'll hear back and see if the problem was the problem and if the fix was the fix. Assembly has been cruising along and more or less all units are prepared for shipment, we just need to see what happens from the problem unit. I'll let you guys know.

See you all later.

Share this post


Link to post
Share on other sites

Congratulations Ben!


customer server+AudiophileOptimizer >>UltraRendu (SR4) >> Lush(JSSG360) >>> IsoRegen(SR4) >>> Lush^2 >>> blu2 >>Blaxius^2D >> Dave > HD800(SDRmod)

Share this post


Link to post
Share on other sites

Ben,

 

Congrads on the wedding! Enjoy!

 

I would guess that implies an official single config spouse (No 'jplay' dual configs for you) from now on.... Or does Chinese law cut you some slack there..? :-)


Bits to analog: HP Pro Server [Win10Pro, software raid; Roon Core] -> Metrum Ambre -> BADA [series1]

Analog to speaks: ARC Ref3 -> ARC DS450M's -> Dynaudio C1 Sig's + 2 REL G2's

Share this post


Link to post
Share on other sites

Congratulations Ben!

to keep you informed, _I just received an email with my tracking number.


Server: CAPS Zuma, SOtM tX-USBexp PCI, JCAT SATA & Battery for OS SSD

SW: Win 8, JRiver+JPlay+Dirac Live, Fidelizer, CAD Script

PSU: Teradak ATX 210 LPS, Paul Hynes SR7EHD 2 Rails

DDC: Berkeley Audio Alpha USB, Genesis Digital Lens (AT&T Fiber Out)

DAC: Wadia 2000 Digimaster (Sledgehammer Buffer, AT&T Fiber In)

Tube Amplifier: Leben CS-600 (NOS KT66 GEC)

Loudspeakers: Claravox Euritmica

Power Conditioners: Accuphase PS-510, Einstein NF6020, Kemp Power Source+, Kemp Power Station 75, Clearaudio APG, Gate IR-1

Share this post


Link to post
Share on other sites

I still would like to know what was the problem with some units as it is my equipment at stake which potentially is at risk.

 

Also once we know what was the problem we can confirm that our units are ok.

 

Cheers

Share this post


Link to post
Share on other sites
Congratulations Ben!

to keep you informed, _I just received an email with my tracking number.

 

Has anybody else received their tracking number yet?


customer server+AudiophileOptimizer >>UltraRendu (SR4) >> Lush(JSSG360) >>> IsoRegen(SR4) >>> Lush^2 >>> blu2 >>Blaxius^2D >> Dave > HD800(SDRmod)

Share this post


Link to post
Share on other sites

Hi All,

 

Let me try to provide some information on the 'problem' and the 'fix'. In case you wonder I'm the guy raising the issue with Teradak after I received my unit.

 

First of all, the original unit works. After receiving it I setup it up and my PC boot without any issue. I ran it for a while and then I noticed that the 5V and 3.3V fluctuates a bit using the h/w monitor provided by my ASRock motherboard. I then proceeded to measure the ripple using my DMM and found that the 12V rail is fine but the 5V rail has ripple at ~100mV which IMO is higher than desirable. I didn't measure the 3.3V rail as I cannot find an easy measuring point with the PC running.

 

Teradak's engineer could not reproduce my observation but nonetheless they investigated further and did confirm that the 5V/3.3V ripples are higher than expected. After some more investigation Teradak developed a modification aimed to improve voltage stability and reduce ripple.

 

The modification involves replacing the voltage reference semiconductor (for each voltage rail) with a green LED (which has constant voltage drop so can serve as voltage reference). Teradak sent me the parts and I did the modification to my PSU. I measured the PSU's performance before and after the modification using a scope with the following results:

 

[12V Rail]

- Original ripple RMS: ~40mV

- LED-mod ripple RMS: ~40mV

 

 

[5V Rail]

- Original ripple RMS: ~110mV

- LED-mod ripple RMS: ~25mV

 

 

So it is clear that the 5V ripple performance improved significantly. Again I didn't measure the 3.3V rail but I believe it will have similar improvements. The voltage stability of all three rails are now very stable as well. I believe Teradak applied the mod to all unshipped units. If your unit has green glow inside the chassis then it has the LED mod.

 

Have fun with your new ATX-PSU.

Share this post


Link to post
Share on other sites
Hi All,

 

Let me try to provide some information on the 'problem' and the 'fix'. In case you wonder I'm the guy raising the issue with Teradak after I received my unit.

 

First of all, the original unit works. After receiving it I setup it up and my PC boot without any issue. I ran it for a while and then I noticed that the 5V and 3.3V fluctuates a bit using the h/w monitor provided by my ASRock motherboard. I then proceeded to measure the ripple using my DMM and found that the 12V rail is fine but the 5V rail has ripple at ~100mV which IMO is higher than desirable. I didn't measure the 3.3V rail as I cannot find an easy measuring point with the PC running.

 

Teradak's engineer could not reproduce my observation but nonetheless they investigated further and did confirm that the 5V/3.3V ripples are higher than expected. After some more investigation Teradak developed a modification aimed to improve voltage stability and reduce ripple.

 

The modification involves replacing the voltage reference semiconductor (for each voltage rail) with a green LED (which has constant voltage drop so can serve as voltage reference). Teradak sent me the parts and I did the modification to my PSU. I measured the PSU's performance before and after the modification using a scope with the following results:

 

[12V Rail]

- Original ripple RMS: ~40mV

- LED-mod ripple RMS: ~40mV

 

 

[5V Rail]

- Original ripple RMS: ~110mV

- LED-mod ripple RMS: ~25mV

 

 

So it is clear that the 5V ripple performance improved significantly. Again I didn't measure the 3.3V rail but I believe it will have similar improvements. The voltage stability of all three rails are now very stable as well. I believe Teradak applied the mod to all unshipped units. If your unit has green glow inside the chassis then it has the LED mod.

 

Have fun with your new ATX-PSU.

 

Was it to replace the voltage references?

 

Can you please post details?


fmak

Share this post


Link to post
Share on other sites

> Was it to replace the voltage references?

 

Yes.

> Can you please post details?

 

I am not sure if you have contacted Teradak but I think it's better for you to ask them to send you the parts and the detailed instructions of the mod. The mod is not difficult if you're handy with a soldering iron.

 

 

Share this post


Link to post
Share on other sites

Thanks, _JL_ for clarifying what was the problem.

 

 

[12V Rail]

- Original ripple RMS: ~40mV

- LED-mod ripple RMS: ~40mV

 

 

[5V Rail]

- Original ripple RMS: ~110mV

- LED-mod ripple RMS: ~25mV

 

Those numbers are nowhere near to what was advertised to us on page 3 I think.

 

I have gotten a reply for the ripple/noise readings on the ATX 275. Michael told me that all 3 ATX PSUs run at similar levels, so we went with the middle of the road PSU for the testing to get some numbers.

 

[ATTACH=CONFIG]14886[/ATTACH]

 

Share this post


Link to post
Share on other sites

Even with a Green LED voltage reference voltage modification,the figures are more than 10 x worse than the figures given in post 60 on P.3. YES, it DOES matter ! The lower the noise, the higher the SQ. The original figures were good, but not great.


How a Digital Audio file sounds, or a Digital Video file looks, is governed to a large extent by the Power Supply area. All that Identical Checksums gives is the possibility of REGENERATING the file to close to that of the original file.

 

PROFILE UPDATED 18-06-2019

Share this post


Link to post
Share on other sites

I think I need to clarify the ripple measurements. Manufacturer's measurements are almost always done with a resistive load.

 

My measurements in contrast are done at the motherboard end with the ASRock motherboard running. Thus noises generated by the motherboard as well as fluctuating current draws by the motherboard/HDD/etc., all adds to the measured noise. So to be fair the noise figures I measured are a sum of the PSU noise plus noises generated by the motherboard.

 

To get a fair comparison one will need to setup suitable resistors as the load at the given current to do the measurements. On an absolute term the ripple performance is respectable IMO. Getting down to a few mV with a running motherboard is next to impossible IMO.

Share this post


Link to post
Share on other sites
I think I need to clarify the ripple measurements. Manufacturer's measurements are almost always done with a resistive load.

 

My measurements in contrast are done at the motherboard end with the ASRock motherboard running. Thus noises generated by the motherboard as well as fluctuating current draws by the motherboard/HDD/etc., all adds to the measured noise. So to be fair the noise figures I measured are a sum of the PSU noise plus noises generated by the motherboard.

 

To get a fair comparison one will need to setup suitable resistors as the load at the given current to do the measurements. On an absolute term the ripple performance is respectable IMO. Getting down to a few mV with a running motherboard is next to impossible IMO.

 

I agree. One cannot compare these two sets of measurements IMO.


Server: CAPS Zuma, SOtM tX-USBexp PCI, JCAT SATA & Battery for OS SSD

SW: Win 8, JRiver+JPlay+Dirac Live, Fidelizer, CAD Script

PSU: Teradak ATX 210 LPS, Paul Hynes SR7EHD 2 Rails

DDC: Berkeley Audio Alpha USB, Genesis Digital Lens (AT&T Fiber Out)

DAC: Wadia 2000 Digimaster (Sledgehammer Buffer, AT&T Fiber In)

Tube Amplifier: Leben CS-600 (NOS KT66 GEC)

Loudspeakers: Claravox Euritmica

Power Conditioners: Accuphase PS-510, Einstein NF6020, Kemp Power Source+, Kemp Power Station 75, Clearaudio APG, Gate IR-1

Share this post


Link to post
Share on other sites

Hi All,

 

I just received another update from Teradak. They further refined the modification with an additional capacitor added to each voltage rail. According to Teradak the ripple performance is now at several mV level when connected to an AMD motherboard drawing >2A per rail:

 

3.3V=1.296mV(RMS)

5V=2.350mV(RMS)

12V=2.998mV(RMS)

 

These photos are from Teradak.

 

IMG_0385.jpg

 

Teradak also informed me that only the first 4 units (1 for fmak, 2 for Ben, and mine) were shipped with the non-mod circuit. All the others have been updated with both the LED mode and the capacitor mod.

IMG_0383.jpg

IMG_0384.jpg

Share this post


Link to post
Share on other sites

_JL_ thank you one more time for another update.

 

Now those numbers looks much better :)

 

Do you know, by any chances, if they shipped all of them already or is it still work in progress?

 

Cheers

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×
×
  • Create New...