Jump to content
IGNORED

Discussion of LiFePO4 battery mods of REGEN and other out-of-warranty adventures


Recommended Posts

[i could not figure out how to get the below note to appear at the beginning of this thread--created by moving posts from another--so I am just editing into the first post on the topic.]

 

 

Several members suggested that I move recent discussions of people modifying their REGENs (to power the hub chip directly--with a 3.2V lithium battery--bypassing the REGEN's already ultra-low-noise regulator) to a new thread so as not to confuse or divert the majority, who are non-DIY hackers.

 

So here are all the past few day's posts on the topic, moved from the thread Uptone Audio Regen .

 

Cheers,

 

--Alex C.

UpTone Audio

 

 

 

 

To Superdad: have you seen the discussion on Audio Asylum about the use of LIFePo4 batteries to power the 3.3 volt circuits on the Regen mobo? The sonics are fantastic. If your new power supply can beat this it will be worth it.

Link to comment
To Superdad: have you seen the discussion on Audio Asylum about the use of LIFePo4 batteries to power the 3.3 volt circuits on the Regen mobo? The sonics are fantastic. If your new power supply can beat this it will be worth it.

 

A link to the discussion would be nice.

Link to comment

"What my experiments showed was that the voltage regulator is probably a limiting factor. Running the Regen directly from LiFePo4 (bypassing the Vreg) was substantially better sounding than powering the Regen using LiFePO4 batteries powering the Regen through its DC input socket.

 

And I didn't find much notable difference between SMPS, LPS or batteries when powering through its DC input socket."

 

Quoted Jkenny from the CAA site.

 

He has a thread on where and how to do this.

http://www.audioasylum.com/cgi/vt.mpl?f=pcaudio&m=153125

(JRiver) Jetway barebones NUC (mod 3 sCLK-EX, Cybershaft OP 14)  (PH SR7) => mini pcie adapter to PCIe 1X => tXUSBexp PCIe card (mod sCLK-EX) (PH SR7) => (USPCB) Chord DAVE => Omega Super 8XRS/REL t5i  (All powered thru Topaz Isolation Transformer)

Link to comment

I think you have to separate out the two areas of improvement that the Regen addresses - the 5V supply & the improvement of the digital signal in any reporting. I believe those who don't find much differences between Regen power supplies are using DACs that don't use the USB 5V & therefore any improvements noticed are due to the USB signal re-conditioning and/or noise floor differences.

 

What's being said above in the quote by jkenny is clearly only related to the 3.3V powered part of the Regen i.e. the USB signal re-conditioning

Link to comment
I think you have to separate out the two areas of improvement that the Regen addresses - the 5V supply & the improvement of the digital signal in any reporting. I believe those who don't find much differences between Regen power supplies are using DACs that don't use the USB 5V & therefore any improvements noticed are due to the USB signal re-conditioning and/or noise floor differences.

 

What's being said above in the quote by jkenny is clearly only related to the 3.3V powered part of the Regen i.e. the USB signal re-conditioning

 

My DAC doesn't use the +5v, and I hear I significant (to me) increase in clarity and detail just from a cheap TeraDak U9 LPS. I'd really like to try just powering the 3.3v with the LPS, but I'm not all that interested in opening up my Regen.

Link to comment
I think you have to separate out the two areas of improvement that the Regen addresses - the 5V supply & the improvement of the digital signal in any reporting. I believe those who don't find much differences between Regen power supplies are using DACs that don't use the USB 5V & therefore any improvements noticed are due to the USB signal re-conditioning and/or noise floor differences.

 

What's being said above in the quote by jkenny is clearly only related to the 3.3V powered part of the Regen i.e. the USB signal re-conditioning

 

I don't believe your generalization is completely accurate. The QB-9 DSD does not use USB 5V - there was a clear improvement switching from the provided SMPS to a JS-2 powering the Regen.

Main System: Mac mini (Audirvana+, MMK, JS-2) -> ISO Regen (LPS-1) -> Icron 2201 (Rex LPS-1.2) -> ISO Regen (LPS-1.2) -> Ayre QB-9 Twenty -> Headamp GS-X Mk2 -> Classe CT-M600 -> KEF Reference 201/2

 

Link to comment

Well, I should have said that there are also many variables in this so it's quite difficult to generalise

- the quality of the USB signal from the source

- the USB cables used

- the DAC used & it's ability to deal with noise on the incoming grounds

 

 

But there's no doubt that both the 5V USB supply & the USB signal are both addressed in the Regen so it makes it difficult to read reports of improvements & where the improvement is coming from - better quality 5V USB supply or better quality USB signal?

 

What I was pointing out was that the USB signal quality was being specifically addressed in the above quote.

Link to comment
I'll add that upgrading the SMPS to the El Cheapo 9v r-core linear supply made an audible improvement. Perhaps it's expectation bias, but if I think it sounds better, then it's money well spent, regardless.

 

Using a Super Regulator, which gives a further improvement in noise levels over a cheap LM317T I.C. based power supply, as well as a lower output impedance, would also appear to help to negate the problems noted by John Kenny re the deficiencies of the already better than average installed 3.3V regulator.

Uptone's new "Mystery" PSU seems likely to confer similar benefits for many Regen users, with possibly a further small improvement over even a JS2 supply which is shared with other areas. The better the power going into the Regen's internal 3.3V regulator, the less work it has to do, other than provide a very stable output voltage.

 

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 13-11-2020

Link to comment
Using a Super Regulator, which gives a further improvement in noise levels over a cheap LM317T I.C. based power supply, as well as a lower output impedance, would also appear to help to negate the problems noted by John Kenny re the deficiencies of the already better than average installed 3.3V regulator.

Uptone's new "Mystery" PSU seems likely to confer similar benefits for many Regen users, with possibly a further small improvement over even a JS2 supply which is shared with other areas. The better the power going into the Regen's internal 3.3V regulator, the less work it has to do, other than provide a very stable output voltage.

 

Don't think so - the quote specifically says that "bypassing the 3.3V regulator" was significantly better than providing even the best, lowest noise power to the DC input of the Regen

Link to comment
Don't think so - the quote specifically says that "bypassing the 3.3V regulator" was significantly better than providing even the best, lowest noise power to the DC input of the Regen

 

Unlike John ,I have obtained improvements by going from the original +7.5V SMPS plugpack to a 12V SMPS plugpack regulated to +9V and followed by a John Linsley Hood PSU add-on, a further small improvement by going to a +9V Linear PSU followed by a JLH, and a further small improvement when going to the original +12V SMPS version, but plugging in a 3,000mAh Li Ion battery instead of the +12V SMPS.

 

P.S.

I have a great deal of respect for John Kenny, as we have worked together on many previous occasions.

 

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 13-11-2020

Link to comment
Don't think so - the quote specifically says that "bypassing the 3.3V regulator" was significantly better than providing even the best, lowest noise power to the DC input of the Regen

 

I agree. We all know that different levels of SQ can be achieved thru various sources using the current 3.3V regulator on the Regen. The real question now is, Does the bypassing of the 3.3V regulator give significantly better SQ than any of those other sources with the 3.3V regulator on the Regen?

I would like to give it a try. Does anyone have a specific recommendation for a trickle charger (power regulator?) to be permanently wired to the following 3.2V LiFePo4 battery?

 

~http://www.ebay.com/itm/320657847254?_trksid=p2060353.m1438.l2649&ssPageName=STRK%3AMEBIDX%3AIT

(JRiver) Jetway barebones NUC (mod 3 sCLK-EX, Cybershaft OP 14)  (PH SR7) => mini pcie adapter to PCIe 1X => tXUSBexp PCIe card (mod sCLK-EX) (PH SR7) => (USPCB) Chord DAVE => Omega Super 8XRS/REL t5i  (All powered thru Topaz Isolation Transformer)

Link to comment

I am thinking you could use the existing 3.3v regulator as a charger for the batteries in the thread.

I agree. We all know that different levels of SQ can be achieved thru various sources using the current 3.3V regulator on the Regen. The real question now is, Does the bypassing of the 3.3V regulator give significantly better SQ than any of those other sources with the 3.3V regulator on the Regen?

I would like to give it a try. Does anyone have a specific recommendation for a trickle charger (power regulator?) to be permanently wired to the following 3.2V LiFePo4 battery?

 

~http://www.ebay.com/itm/320657847254?_trksid=p2060353.m1438.l2649&ssPageName=STRK%3AMEBIDX%3AIT

Forrest:

Win10 i9 9900KS/GTX1060 HQPlayer4>Win10 NAA

DSD>Pavel's DSC2.6>Bent Audio TAP>

Parasound JC1>"Naked" Quad ESL63/Tannoy PS350B subs<100Hz

Link to comment
Unlike John ,I have obtained improvements by going from the original +7.5V SMPS plugpack to a 12V SMPS plugpack regulated to +9V and followed by a John Linsley Hood PSU add-on, a further small improvement by going to a +9V Linear PSU followed by a JLH, and a further small improvement when going to the original +12V SMPS version, but plugging in a 3,000mAh Li Ion battery instead of the +12V SMPS.

 

P.S.

I have a great deal of respect for John Kenny, as we have worked together on many previous occasions.

Yes, I guess it does depend on the many factors I mentioned already like the sensitivity of the DAC to noise borne on the USB ground or signal wires.

I know you have taken this into account so don't get upset at me asking but I presume the DAC you used was not using 5V from the Regen?

Link to comment
Yes, I guess it does depend on the many factors I mentioned already like the sensitivity of the DAC to noise borne on the USB ground or signal wires.

I know you have taken this into account so don't get upset at me asking but I presume the DAC you used was not using 5V from the Regen?

 

My findings were mainly based on using the Regen and PSU with a Corsair Voyager into the USB input of an Oppo 103.

The earlier reports were based on using the Regen into the USB input of a Bricasti M1 DAC which does draw some current from the Regen. I hope to be able to measure that current draw at our next listening session, using a cheap inline Current and Voltage meter gizmo.

Alex

 

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 13-11-2020

Link to comment
My findings were mainly based on using the Regen and PSU with a Corsair Voyager into the USB input of an Oppo 103.

The earlier reports were based on using the Regen into the USB input of a Bricasti M1 DAC which does draw some current from the Regen. I hope to be able to measure that current draw at our next listening session, using a cheap inline Current and Voltage meter gizmo.

Alex

 

Ok, glad I asked - the quality of the 5V comes into play in a big way in the two scenarios you mention so it's not really surprising that the PS feeding the Regen makes a noticeable difference in these configurations?

 

This is what I was trying to tease out with my posts - improvements due to the 5V conditioning of the Regen Vs improvements due to conditioning of the USB signal

Link to comment
I am thinking you could use the existing 3.3v regulator as a charger for the batteries in the thread.

Yes, this is a good idea, not quite sure how one would go about it. The other concern is the following from JKenny. I don't need the 5V.

 

"You can try using the 3.3V regulator as a trickle charger for the battery but remember powering the Regen from the DC input will also power the 5V regulator which may put some noise on the ground plane."

(JRiver) Jetway barebones NUC (mod 3 sCLK-EX, Cybershaft OP 14)  (PH SR7) => mini pcie adapter to PCIe 1X => tXUSBexp PCIe card (mod sCLK-EX) (PH SR7) => (USPCB) Chord DAVE => Omega Super 8XRS/REL t5i  (All powered thru Topaz Isolation Transformer)

Link to comment

I was thinking something like this switching power supply would work for recharging the battery.

 

http://www.ebay.com/itm/351455924597?_trksid=p2060353.m1438.l2649&ssPageName=STRK%3AMEBIDX%3AIT

(JRiver) Jetway barebones NUC (mod 3 sCLK-EX, Cybershaft OP 14)  (PH SR7) => mini pcie adapter to PCIe 1X => tXUSBexp PCIe card (mod sCLK-EX) (PH SR7) => (USPCB) Chord DAVE => Omega Super 8XRS/REL t5i  (All powered thru Topaz Isolation Transformer)

Link to comment

You simply plug in the Regen power to charge the battery, and unplug it to listen via battery. It really couldn't get easier than that. Alternately you may leave it plugged in for a constant trickle charge with the aforementioned shared ground.

Yes, this is a good idea, not quite sure how one would go about it. The other concern is the following from JKenny. I don't need the 5V.

 

"You can try using the 3.3V regulator as a trickle charger for the battery but remember powering the Regen from the DC input will also power the 5V regulator which may put some noise on the ground plane."

Forrest:

Win10 i9 9900KS/GTX1060 HQPlayer4>Win10 NAA

DSD>Pavel's DSC2.6>Bent Audio TAP>

Parasound JC1>"Naked" Quad ESL63/Tannoy PS350B subs<100Hz

Link to comment

Regarding the John Kenny battery mod discussion:

 

I am thinking you could use the existing 3.3v regulator as a charger for the batteries in the thread.

 

I was thinking something like this switching power supply would work for recharging the battery.

 

US DC 3 3V 1000mA Switching Power Supply Adapter 100 240 AC 50 60Hz | eBay

 

You simply plug in the Regen power to charge the battery, and unplug it to listen via battery. It really couldn't get easier than that. Alternately you may leave it plugged in for a constant trickle charge with the aforementioned shared ground.

 

John Kenny's battery mod requires that his recommended 22650 batteries be plugged into the 3.3V regulated supply during play. You must not unplug the charger during play.

 

2X 26650 3 2V 3 3Ah LiFePO4 Rechargeable Battery w Tabs | eBay

 

Looking at the specs, the batteries were designed to be charged with a 3.65V charger and many discharge curves can be found online showing that 3.2V LiFeP04 batteries, when fully charged, can deliver 3.4V or higher at the beginning of their discharge cycle. That's why John Kenny suggests using a 3.3V regulated power supply instead of a typical 22650 charger.

 

The USB Regen doesn't want to see 3.65V (as it would if the battery was plugged into a typical 22650 charger). The USB Regen requires a regulated 3.3V.

 

The specs also show that this 22650 battery pack has a nominal rating of 3.2V - meaning that the average voltage throughout most of its discharge cycle will be 3.2V.

 

The USB Regen doesn't want to see anything less than 3.3V (as it would throughout most of the battery's discharge cycle were you to leave the 3.3V regulated power supply unplugged during play.) The USB Regen requires a regulated 3.3V.

 

So, even if you charge the battery with a 3.3V regulated supply instead of a 3.65V charger, you cannot disconnect the battery during play, because the voltage will drop with use.

 

HTB1ImpwJpXXXXbnXFXXq6xXFXXXL.jpg

 

That's the gimmick with John Kenny's mod: He keeps the voltage of the battery pack at a constant 3.3V during play by allowing it to stay plugged into a regulated 3.3V power supply.

 

He never allows the battery pack to come to a full charge (>3.5V)

He never allows the battery pack to discharge.

He keeps the battery pack at a constant 3.3V (as required by the USB Regen).

 

Disclaimer: I personally have no intention of trying John Kenny's mod and I am not recommending it, here. I'm just trying to do a better job of explaining it than he has in his Audio Asylum threads.

Link to comment
You simply plug in the Regen power to charge the battery, and unplug it to listen via battery. It really couldn't get easier than that. Alternately you may leave it plugged in for a constant trickle charge with the aforementioned shared ground.

 

Yes, 4est, great idea. I like it. I wonder how long the charge would hold? Zilch may have a point in keeping constant power attached to the rechargeable battery. But I like the simplicity of your idea. My listening sessions usually never last more than 6 hours.

(JRiver) Jetway barebones NUC (mod 3 sCLK-EX, Cybershaft OP 14)  (PH SR7) => mini pcie adapter to PCIe 1X => tXUSBexp PCIe card (mod sCLK-EX) (PH SR7) => (USPCB) Chord DAVE => Omega Super 8XRS/REL t5i  (All powered thru Topaz Isolation Transformer)

Link to comment
Regarding the John Kenny battery mod discussion:

Disclaimer: I personally have no intention of trying John Kenny's mod and I am not recommending it, here. I'm just trying to do a better job of explaining it than he has in his Audio Asylum threads.

I think you have some things correct & over-egging some others

 

LiFePO4 battery chargers have two phases of operation during charging - first phase is constant current & last phase is constant voltage. In the first phase the voltage will rise to whatever is required to fulfill the constant current setting - at a certain point the charger will switch to constant voltage. Battery chargers operate in this way to speed up the recharging.

 

Trickle charging is a different method - it involves no constant current phase, just a constant voltage & is therefore slower. Using this method @ 3.3V will bring a battery to almost fully charged but does also prevent it from an over-voltage of 3.3V. I guess this could be tested by running solely from battery to see how many hours it runs for?

 

Your discharge graph is at 0.5C means that the current being drawn is 0.5 of the capacity i.e for a 3Amp capacity battery this is discharging at 1.5Amps. This is very far above the current draw of the Regen's USB hub chip & clock (which is more like 100mA tops) so I doubt you will see anything like the voltage drop seen in that graph. Furthermore the USB2412 chip datasheet specifies a operating range of 3V to 3.6V & typical clocks have the same voltage range of operation.

 

So I don't see anything wrong with the information on audioasylum when you work out the correct operating conditions

Link to comment
Yes, 4est, great idea. I like it. I wonder how long the charge would hold? Zilch may have a point in keeping constant power attached to the rechargeable battery. But I like the simplicity of your idea. My listening sessions usually never last more than 6 hours.

 

The USB Regen requires a constant 3.3V supply. That small 22650 battery pack, unplugged from John Kenny's recommended 3.3V regulated power supply, cannot supply 3.3V to the USB Regen continuously across six hours of discharge, without dropping in voltage. It's unlikely the battery can maintain the 3.3V required by the USB Regen for even one hour of discharge.

Link to comment
I think you have some things correct & over-egging some others

 

LiFePO4 battery chargers have two phases of operation during charging - first phase is constant current & last phase is constant voltage. In the first phase the voltage will rise to whatever is required to fulfill the constant current setting - at a certain point the charger will switch to constant voltage. Battery chargers operate in this way to speed up the recharging.

 

Trickle charging is a different method - it involves no constant current phase, just a constant voltage & is therefore slower. Using this method @ 3.3V will bring a battery to almost fully charged but does also prevent it from an over-voltage of 3.3V. I guess this could be tested by running solely from battery to see how many hours it runs for?

 

Your discharge graph is at 0.5C means that the current being drawn is 0.5 of the capacity i.e for a 3Amp capacity battery this is discharging at 1.5Amps. This is very far above the current draw of the Regen's USB hub chip & clock (which is more like 100mA tops) so I doubt you will see anything like the voltage drop seen in that graph. Furthermore the USB2412 chip datasheet specifies a operating range of 3V to 3.6V & typical clocks have the same voltage range of operation.

 

So I don't see anything wrong with the information on audioasylum when you work out the correct operating conditions

 

OK, thanks for that.

 

If the USB Regen can truly be happy with an otherwise unregulated voltage that falls anywhere in the range of 3.0 V to 3.6V, everyone can just relax and use the battery pack without keeping it connected to a regulated 3.3V supply - a departure from John Kenney's recommendation - as long as they somehow monitor the battery voltage during play to ensure that it never falls below 3.0V.

 

But why not do as John Kenney recommends - keeping the battery connected to a regulated 3.3V supply during play, so that it is never supplying the USB Regen with a voltage any higher or lower than 3.3V?

Link to comment

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



×
×
  • Create New...