UltraCap™ LPS-1 Operation and Pre-purchase thread

[JohnSwenson]

I have what I hope is a fairly simple question to answer. I recall reading somewhere that it is best to leave the LPS-1 'powered up' permanently, along with say it's associated microRendu or whatever. OK - fine, I understand this, and a proportion of the kit I have is left on permanently.

 

However, I recall reading somewhere that the LPS-1 (or ultracaps in general?) do not like being turned off much, or at least they get hotter when turned off, prior to then cooling of course. The point being, I do quite regularly have periods where I know I will not be able to use my system maybe for a week, sometimes two weeks. If an enforced two weeks away from the system is approaching, it does make sense to turn everything off. Does this have any kind of negative effect on the LPS-1? Would the recommendation be keep everything powered up regardless? I suspect this may be a non issue, but I would be interested if there is any good advice here.

 

Turning the LPS-1 off occasionally does not harm anything. Turning it off once a day is fine, turning it off twenty times a day is probably not a good thing to do. Putting the electronics through that much thermal cycling is not good, let alone that your audio gear will never sound its best since it will always be cooling off and warming up.

 

John S.

[JohnSwenson]

Owners of the LPS-1: HOW WARM DOES YOUR UNIT RUN?

 

I ask because I'm wanting one, but am reluctant to add another very warm item to my listen room. May sound silly, but without AC that room gets warm with all my equipment on in the summer.

 

I plan to use the Meanwell PSU to begin with, but I also own a JS-2 and could use one of its variable rails to power the LPS-1 powering my uRendu. (My DAC is an Exogal Comet.)

 

Does the warmth of the LPS-1 depend on the input voltage? And if so, could I keep the heat down to, say, the lukewarm feeling of my uRendu by selecting the optimal voltage? (SQ is constant, I understand.)

 

I've seen one or two reports of the LPS-1 running very warm, and that's the root of my first-world concern.

 

Thanks!

The power dissipation of the LPS-1 depends almost entirely on the output current only, not the input voltage or the output voltage. At 1A output current it comes out to around 3.8W. Not that much heat.

 

Of course the actual temperature of the case will depend on the ambient temperature, the airflow, orientation etc.

 

Note that the amount of heat going into the room does not depend on the temperature of the case, it is the power dissipated by the electronics, which is worst case less than 4 watts.

 

John S.

[JohnSwenson]

Hi,

 

My LPS-1 is on its way. I am currently using a MCRU LPS 7V DC to power my microRendu. When my LPS-1 arrives, can I use the MCRU unit to power it or will it be insufficient? It seems that I need 7.5V/2.5A, so just short. Will it do any harm if I try it anyway?

 

I have ordered the compatible energising supply but, given that the MCRU is already in situ, I would like to use it if I'm able.

 

A 7V supply WILL work as the feeder for an LPS-1 as long as it has a very low output impedance. At 7V the LPS-1 will be taking close to 3A (somewhere around 2.9A), if at 2.9A the feeder drops very much from 7V it is not going to work. Even though 7V will work it has to be VERY well regulated and a short low resistance cable between the feeder and the LPS-1.

 

A kind of runaway feedback can happen. As the feeder voltage gets lower the LPS-1 input current goes up, which can cause the voltage to get lower, which causes the current to get higher ... This is more of a problem at lower voltages because the current is higher to begin with. As long as the feeder can handle the high currents with very little droop this will not happen.

 

You can find a hint about how well a given 7V supply will do by looking at it's current rating. If it says anything less than 3A, completely forget it. If it is right at 3A it should theoretically work, but most 3A supplies will have too much droop. If it is significantly over 3A, say 5A, it has a decent probability of working (it may still have too much droop).

 

All the above is why the official spec only goes down to 7.5V, lower voltages MAY work, but it takes an exceptional supply to do so.

 

John S.

[JohnSwenson]

Here is my recommendations for microRendu, LPS-1 and DACs.

 

If your DAC is VBUS powered only (ie no external power jack), plug it into the microRendu USB jack and power the microRendu from the LPS-1. Use a good short USB cable or a good solid adapter. There is no reason for cutting grounds etc in this configuration, cutting the ground blocks leakage currents, with the LPS-1 you don't have any so it is unnecessary. I have designed a really nice solid adapter that will work great for this, but Alex hasn't started making them yet, too busy with the LPS-1.

 

If the DAC has an external power port, and will run at 7V and takes 1/2A or less, you CAN use a Y cable from an LPS-1 to drive both the microRendu and the DAC. This is the way I have my setup and it works VERY well.

 

It is certainly possible to use a separate LPS-1 for both the microRendu AND the DAC, It MAY sound better than a single driving both, but I cannot guarantee that, it is going to be very system dependent. It should not sound worse, but it may not be any better, or it may be a big improvement, hard to tell.

 

Using a USB power injector to feed power into a VBUS only powered DAC, should not be necessary if you have a good short USB cable.

 

Whatever you do, do not use a USB power injector with something other than an LPS-1 if you are using an LPS-1 to power the microRendu. This will nullify some of the advantages of using the LPS-1 with the microRendu. Put another way, IF you have a DAC that can be powered from an LPS-1, power it from either its own LPS-1, or the same one that is powering the microRendu IF the DAC current is low enough to allow that.

 

If your DAC has it's own built in AC supply obviously you cannot power it from an LPS-1, in this case still power the microRendu from the LPS-1, it will still be a large improvement, but not quite as much as you would get if the DAC was powered off an LPS-1 as well.

 

John S.

[JohnSwenson]

 

Now I will not speak for John or Sonore with regards to technically acceptable operating voltage range, but prompted by your post I set my LPS-1 to 5V and connected the MicroRendu. It seems happy--even running into my non-VBUS-using DAC.

 

The microRendu has a 5V regulator that drives several other lower voltage regulators. If you feed 5V in, that regulator will be out of regulation outputting somewhat less than 5V to the other regulators. This will work, but you are giving up on the regulation of that first regulator which is part of the power supply isolation scheme in the microRendu. So while it will actually work, it is not recommended.

 

John S.

[JohnSwenson]

I take it the same would apply to Regen/DAC?

 

Things get a little trickier with a REGEN. A microRendu powered by an LPS-1 has no leakage current to pass down the USB cable, there is none from the PS and the Ethernet connection blocks the leakage current from the digital side of things.

 

A REGEN does NOT completely block the leakage current from the computer, so with an LPS-1 powering a REGEN you still can have a leakage loop from the computer PS. You block a leakage loop going through the power supply of the REGEN, but the loop from the computer through the USB cable is still there.

 

Powering the REGEN with an LPS-1 is still going to sound better than powering the REGEN from the MeanWell, but exactly how to arrange things may be different once you include the loop through the computer.

 

John S.

[JohnSwenson]

There has been a lot of discussion about using Y cables to power multiple devices from the LPS-1 output. What about the input side? In other words, if I have two LPS-1s, can I power them from a single power supply with a Y cable? What is the minimum size supply that can handle two?

 

Yes you can power two LPS-1 from the same power supply. Take the max current draw of one LPS-1 at whatever voltage your feed supply outputs and multiply by two. Your feed supply has to be able to support that.

 

Remember that at the lower voltage end of the scale that is going to be a LOT of current. You will need to make SURE that your Y cable is made from low resistance wire (ie 18 AWG wire may not be good enough).

 

The two LPS-1 outputs are still fully isolated from each other.

 

John S.

[JohnSwenson]

I have had my LPS-1 for a couple of weeks now, I shall be posting my listening observations in the appropriate thread, but suffice to say here, an improvement in sound quality beyond my expectations, so all good!

 

I am using the LPS-1 to power a microRendu, which is then connected to a Mutec MC3+USB. This is significant as the Mutec does draw current to power it's USB input, and thus isolate the USB for the Mutec itself. There was some debate a while back as to if the LPS-1 could satisfactorily power both the mR and the Mutec's USB due to the 'theoretical' combined current draw. All I can say is that the combination operates perfectly, proof of concept, it works.

 

I do have some practical observations. With the 9V ifi powering the mR+Mutec, the mR used to get staggeringly hot, not good. So I needed to provide some additional cooling for the mR for this set-up to work satisfactorily. With the 7V LPS-1 powering the mR+Mutec, the mR has a nice gently warmth to the touch, no additional cooling needed, a much better result.

 

However, powering the mR + Mutec's USB the LPS-1 itself does get hot. Too hot? No, I would not say so, you can hold your hand on the case for an extended period without burning, It just maybe feels like your hand would slowly cook and tenderise if you left it there for a few hours, so hot enough, but I have had kit in the past that has run a lot hotter, so nothing to be concerned about I think. This raises a question. If I know I will not be using my system for a week or so, I can turn off the Mutec, leaving the LPS-1 powering the mR 24/7. With the Mutec turned off, both the LPS-1 and mR are just slightly warm, no issues at all. So here is my question, lets say I knew I would be away from my system, say for one week. Would I be better off leaving the mR/Mutec/LPS-1 powered up, with the LPS-1 permanently in it's quite hot state, or would the recommendation be to turn off the Mutec, allowing the LPS-1 to revert to the slightly warm state. Any recommendations?

 

In your configuration I think the oscillator in the Mutec is the master clock, so I would just leave everything on all the time to make sure the master clock stays stable.

 

John S.

[JohnSwenson]

That may be a a very good option, as I found out an Y-split was not ideal. And yes a second LPS-1 is the ultimate solution. (I will soon test if an EMO EN-70HD can be sufficient instead of FMC's)

 

What I want to ask Superdad or John Swenson about, is the possibility to make a LPS-1 (MK2?), with two isolated outputs. (Same voltage I suppose). Is that doable or even an option to be considered?

Would they need a fixed max output current?

For those not using a Vbus from the MicroRendu, very little power is needed. So there is a lot of power available for a FMC, an USB DAC or a converter.

 

Or maybe is it more convenient to develop a Y spitter that fulfill the need of isolation between the two (or even 3) sources powered from the LPS-1.

 

And will such a possible solution or option, save enough money?

I'm thinking well below 50% of a LPS-1.

 

The entire circuit of the LPS-1 is what makes it isolated, so in order to have two isolated outputs you need two LPS-1 circuits, the only thing that could possibly be common is the case. Since it would have to be a larger case to fit two boards, it would cost more so there would be VERY little cost savings, maybe $2 at most. NOT worth it!

 

John S.

[JohnSwenson]

10 hours ago, Em2016 said:

Hi guys, is the standard power cable that comes with the LPS-1 shielded?

 

I know it's nothing fancy and cheap but is it at least shielded? If so, that will be fine for me.

 

If not, is the Sonore DC-4 Cardas built power cable shielded?

 

Sorry to ask, but search results on the new forum site haven't been working for me today.

 

Cheers !

 

Yes it is shielded.

 

John S.

[JohnSwenson]

38 minutes ago, Isflicka said:

Is it possible to connect my two LPS-1's for 12 volts

Yes it works very well. You will have to make a reverse Y cable that puts the two outputs in series. Using the techniques in the DIY DC cable thread this is not hard to do.

 

You set one LPS-1 for 5V and the other to 7V.

 

John S.

[JohnSwenson]

2 hours ago, Superdad said:

 

1.6A at 5V? How how does that oven clock get?! (is there room in that oven for microscopic loaves of bread? )

If you 100% certain that the clock really draws that much current, then no, the UltraCap LPS-1 maxes out at between 1.1A-1.2A (we guarantee them to 1.1A, but many test on my bench to 1.15-1.18A) and would not be appropriate.

Have you ever put an ammeter in line with your clock battery supply?  I really would be shocked if it drew anywhere near 1.6A/5V.

 

 

 

That is actually quite common for OCXOs. The oven always takes more current when it is warming up than when it is up to temp and stabilized. Each OCXO has a different point between startup current draw and how long it takes to stabilize. After the oven warms up the current goes way down.

 

An OCXO with 1.6A at 5V will probably stabilize in a couple of minutes. A clock that takes 0.6A at startup might take 5-8 minutes. 

 

One possible solution is to have a DPDT switch and use a "regular" high current supply to swarm up the oven, then switch to the LPS-1 when the current goes down. It could be done with a relay and some form current sensor to switch the relay.

 

John S.

[JohnSwenson]

9 hours ago, No Disc said:

Thinking about getting the LPS-1 in addition to the ISO Regen.  Curious if there is any benefit to powering the LPS-1 via an HD-Plex instead of the mean-well power supply. I believe I read someone else had already done this, but I cannot find the post now.  I'm thinking it cannot hurt, was just looking to see if anyone already did it, and what the result was?

Just be sure that the supply can handle the current as listed on the LPS-1 product page. The energizing current is fixed, it is the same no matter what the output current is. So even if your load is small the "feeder" supply needs to be able to handle the current listed on the product page.

 

Many people have had plugged in their supplies that were supposed to be able to work and found out that their voltage dropped dramatically when feeding the LPS-1, in other words their spec was way over inflated from what the device could really deliver.

 

John S.

[JohnSwenson]

9 hours ago, No Disc said:

I understood what Alex wrote, I was looking for clarification as to why 12V would be less load than 7.5V?   Perhaps I don't really understand how LPS function. 

The LPS-1 input is a switching regulator, this pulls a constant POWER. Thus if you use a higher voltage you need less current, lower voltage, higher current. Most linear supplies don't work this way, if the output takes an amp the input takes an amp. Thus for a linear supply feeding an LPS-1 you want to draw as least amount of current as possible, this happens at 12V, that is the lowest current from the feeder supply.

 

The maximum input voltage is 12V, so DON'T use the variable output set to higher than 12V.

 

John S.

 

 

[JohnSwenson]

The LPS-1 lifespan is very long no matter what you do with it. So leaving it off for two weeks then turning back on is fine, turning it off every day will not significantly shorten or lengthen its lifetime.

 

For the best SOUND you should leave it and what it powers on all the time.

 

John S.

[JohnSwenson]

4 hours ago, Em2016 said:

Hi @JohnSwenson

 

I have two LPS-1's each powering Raspberry Pi's as Roon endpoints around my place. Both running the exact same software and hardware (Dietpi and HiFiBerry Digi+ Pro HAT's).

 

For one of the LPS-1's, when there's no music playing I can hear the cap banks switching, even from a few meters away.

 

For the other LPS-1, I can't hear any switching even with my ear inches from it. For this one, is it possible that the LPS-1 is outputting 5Vdc fine but it's no longer switching cap banks?

 

If the banks were not switching you would not get a sustained current from the output.

 

I have never heard the clicking that you are talking about, but then I have only been in the presence of two production models and 8 prototypes. Alex has been in the presence of hundreds of them and has only heard a couple times, so there is not a lot of experience with it and what causes it etc.

 

So I would say that if you have one that doesn't make any noise, that is the norm, don't worry about it.

 

John S.

[JohnSwenson]

12 hours ago, Em2016 said:

Hi John S @JohnSwenson

 

I've seen you mention that you can power your Bottle Head DAC and microRendu with a single LPS-1 using a Y splitter cable.

 

But you've also said it's not ideal.

 

Since leakage current loops happen between AC power supplies (if I've understood all your great teachings properly), what is the issue with powering a DAC and microRendu with a single LPS-1, if it's not leakage currents that's the issue?

 

I understand the DAC and mR are not isolated if they share the same LPS-1 but if it's not leakage currents that's the issue, what are the other issues? And could this affect the DAC performance?

 

Cheers in advance

You have at least one connector and some cable in common. Almost all digital devices have rapidly changing load currents, these create noise across the impedance of the connectors and wire. Thus the load changes from device A are going to show up as increased noise on the power input to device B.

 

John S.

[JohnSwenson]

2 hours ago, Cyrus said:

@pl_svn  Thank you. Yes, then it will not work in that application.

Does it really take 3A? Many devices take considerably less current than the rating of the supply they come with. If in doubt there are inexpensive USB current meters you can use to actually measure how much it takes. Amazon has about 20 of them for $10.

 

John S.