A few years back batteries were tested for noise vs. frequency and compared to some high performance discrete regulators. I am going to say Marsh did it, but I am guessing since I do not really recall exactly. It may have been a discussion on regulators where Marsh's designs were discussed.
Anyhow, iirc, it was found that the residual noise from batteries in general was higher than the residual noise from the higher performance regulators.
Batteries are not noise free.
Also, one would really have to test this WRT dynamic current and voltage draw vs. frequency to really see what is going on, and which is "better"...
Obviously one could follow a battery with a regulator, etc... but that might obviate the dynamic characteristics (assuming they are desirable in the first place) of the batteries.
If this matters at all depends on factors beyond the scope of my comments.
_-_-bear
Anyhow, iirc, it was found that the residual noise from batteries in general was higher than the residual noise from the higher performance regulators.
Batteries are not noise free.
Also, one would really have to test this WRT dynamic current and voltage draw vs. frequency to really see what is going on, and which is "better"...
Obviously one could follow a battery with a regulator, etc... but that might obviate the dynamic characteristics (assuming they are desirable in the first place) of the batteries.
If this matters at all depends on factors beyond the scope of my comments.
_-_-bear
Where did I repeat it? I said it only once.
But you are right, especially for "SE" topologies.
I misremembered something from years back.
it was TNT audio and ALW, i've already linked it a few pages ago, but while its good to have as reference, it does not include any battery chemistries that are very relevant today
A123 are hard to find... but...
Once A123 stopped selling to small scale race-car aficionados they became very hard to find. If you contact Ping battery (try ebay) you will find they make some cells with extremely low ESR levels that seem comparable to or better than A123 cells, with the added virtue of actually being available.
I think RedWine may be sourcing their raw battery cells from them. The battery management part of the Ping package may be one of the limiting factors on how good the batt-pack sounds, due to limiters on available instant current - translates into more noise and lower instantaneous current.
RedWine designed or modded the battery management circuit specifically for audio applications... and they sure sound good. Expensive? Yes. And they sure sound good.. I keep on coming back to that last line...
I'd love to see some really good DIY type kits for the low ESR Ping flat-pack LiFePo4 cells. It's beyond my current level of interest to delve too far into that aspect, I'd rather do more research on my speakers - but if there are any here who have made better sounding batt management systems, I'd love to see them post on this site!
Once A123 stopped selling to small scale race-car aficionados they became very hard to find. If you contact Ping battery (try ebay) you will find they make some cells with extremely low ESR levels that seem comparable to or better than A123 cells, with the added virtue of actually being available.
I think RedWine may be sourcing their raw battery cells from them. The battery management part of the Ping package may be one of the limiting factors on how good the batt-pack sounds, due to limiters on available instant current - translates into more noise and lower instantaneous current.
RedWine designed or modded the battery management circuit specifically for audio applications... and they sure sound good. Expensive? Yes. And they sure sound good.. I keep on coming back to that last line...
I'd love to see some really good DIY type kits for the low ESR Ping flat-pack LiFePo4 cells. It's beyond my current level of interest to delve too far into that aspect, I'd rather do more research on my speakers - but if there are any here who have made better sounding batt management systems, I'd love to see them post on this site!
Last edited:
Batteries are not noise free?
Nor are resistors noise free. Nothing is. The molecules in the air make noise. And the point is?
But let's put thing in context. The age of the study you referenced? A few years ago? More like 20!! (Or was it 30?)
With LiFePo4, we're discussing new types of batteries that have only been available to the public in the last 3-4 years.
The residual noise of LiFePo4 is extremely low. Heck, I think I should PM Vinnie at RedWine and ask him to contribute to this discussion, there's a lot he could teach us. I know for a fact he uses LiFePo4 straight (with Battery Management) on his 25.6V packs, and uses regulators for some of the lower voltages, such as 5V lines, probably because LiFePo4 is mostly available only in increments of 3.2V per cell.
Nor are resistors noise free. Nothing is. The molecules in the air make noise. And the point is?
But let's put thing in context. The age of the study you referenced? A few years ago? More like 20!! (Or was it 30?)
With LiFePo4, we're discussing new types of batteries that have only been available to the public in the last 3-4 years.
The residual noise of LiFePo4 is extremely low. Heck, I think I should PM Vinnie at RedWine and ask him to contribute to this discussion, there's a lot he could teach us. I know for a fact he uses LiFePo4 straight (with Battery Management) on his 25.6V packs, and uses regulators for some of the lower voltages, such as 5V lines, probably because LiFePo4 is mostly available only in increments of 3.2V per cell.
Well, its nice to know some amps are so well designed that you can hear the difference between batterys with them...
As Jack as just stated a lot of things contribute noise...good design involves minimising the effect this noise has on the signal being processed, amplified etc, bad design means the noise from any source is affecting the sound...
So you evolve your design negate the noise, as stated where I am working on some battery controlled stuff its been an entertainment for the chaps reading this thread.
Cheers
Oh and before you all get a boo boo on, read my earlier thread, I am not saying there are differences between batteries, just like different PSU topographies have different noise spectrums, its the listening part that worries me.......
As Jack as just stated a lot of things contribute noise...good design involves minimising the effect this noise has on the signal being processed, amplified etc, bad design means the noise from any source is affecting the sound...
So you evolve your design negate the noise, as stated where I am working on some battery controlled stuff its been an entertainment for the chaps reading this thread.
Cheers
Oh and before you all get a boo boo on, read my earlier thread, I am not saying there are differences between batteries, just like different PSU topographies have different noise spectrums, its the listening part that worries me.......
Hi Jack,
Thanks for replying & suggesting - & yes it's also my impression that Red Wine gets their batteries from a123systems. And they do indeed seem to be difficult to find although I just found a supplier in Germany on ebay for the larger 20AH cell (and the 2.5AH cell is readily available on ebay):
A123 Systems AMP20M1HD-A, 20Ah Folienzelle, LifePO4 Akku, Battery | eBay
It's good to know about Ping battery as well - just as an alternative, although I seem to read in your post that you yourself are quite taken by the Red Wine battery system and its sound quality ...? Their design, however, will not fit my use - need single cell voltages to be accessible.
Hmmm... delving into battery specifics & BMSs also is not my main focus - I hope to find readily available circuits that I can use for charge & discharge monitoring - while at the same time keeping a direct output from the battery to the circuits I use. Having tinkered with designing for Li-ion batteries before I know this is possible if I accept the risk of a short-circuit. However, since these batteries output so much power in my experience what normally happens is that the components that caused the shortcircuit gets fully "burned" and then "that's it" - slightly mildly said.
Alternatively, a high end audio designer I know once suggested that for fuses I could use a short piece from a transistor leg (I think it was) - his experience was that this sounded ok (fuses apparently not so) and that the leg burned at about ~30Amps. Designing for private use allows for some free space of non-commercial design choices ...
So, with this in mind let me add that the above is not recommended for other users unless you are willing to accept the risk of a fire.
Best regards,
Jesper
Thanks for replying & suggesting - & yes it's also my impression that Red Wine gets their batteries from a123systems. And they do indeed seem to be difficult to find although I just found a supplier in Germany on ebay for the larger 20AH cell (and the 2.5AH cell is readily available on ebay):
A123 Systems AMP20M1HD-A, 20Ah Folienzelle, LifePO4 Akku, Battery | eBay
It's good to know about Ping battery as well - just as an alternative, although I seem to read in your post that you yourself are quite taken by the Red Wine battery system and its sound quality ...? Their design, however, will not fit my use - need single cell voltages to be accessible.
Hmmm... delving into battery specifics & BMSs also is not my main focus - I hope to find readily available circuits that I can use for charge & discharge monitoring - while at the same time keeping a direct output from the battery to the circuits I use. Having tinkered with designing for Li-ion batteries before I know this is possible if I accept the risk of a short-circuit. However, since these batteries output so much power in my experience what normally happens is that the components that caused the shortcircuit gets fully "burned" and then "that's it" - slightly mildly said.
Alternatively, a high end audio designer I know once suggested that for fuses I could use a short piece from a transistor leg (I think it was) - his experience was that this sounded ok (fuses apparently not so) and that the leg burned at about ~30Amps. Designing for private use allows for some free space of non-commercial design choices ...
So, with this in mind let me add that the above is not recommended for other users unless you are willing to accept the risk of a fire.
Best regards,
Jesper
repeating what you have read on the internet.
Although I see you qualified it and now accept it is not true for the majority of ClassA topologies.
"So I've erased myself, replaced my mind, it's a clean slate
All I felt and ev'ryone I've ever known is gone."
(Beardfish - Into The Night)
Well, it just came into my imagination what would it look like if I'd - literally - listen to Li-Ion batteries, as the thread's title tells us ...
Best regards!
Here's battery noise -- the noise floor is with a shorting bar across the balanced input of the analyzer:
The amplifier is an SSM2019 +40dB.
As you can see, the volume of the battery makes it a "radiator" and adds noticeably to the amount of line noise it picks up.
An externally hosted image should be here but it was not working when we last tested it.
The amplifier is an SSM2019 +40dB.
As you can see, the volume of the battery makes it a "radiator" and adds noticeably to the amount of line noise it picks up.
Interesting measurement; it will not only be environmental noise picked up, but also thermal noise. AFAIK NimH can have pretty low internal resistance; lower than lead acid.
I suppose these measurements were made with the batteries not under load. Since (dis)charge is a chemical process, I understand this may be the primary noise generation mechanism in a battery actually doing work. It would be very interesting to see how the noise figures would react to a load being driven.
vac
Hmmmm... interesting. Can I ask you what the capacity is of the two batteries?
Best regards,
Jesper
Jack,
Have you also tried with a 75 Ohm load on the NimH, so that both draw the same current, just to make a fair comparison between the two types?
It might also be interesting to see what happens if both would deliver twice the current by lowering the load resistance. This might bring out more excess noise.
regards, vac
Have you also tried with a 75 Ohm load on the NimH, so that both draw the same current, just to make a fair comparison between the two types?
It might also be interesting to see what happens if both would deliver twice the current by lowering the load resistance. This might bring out more excess noise.
regards, vac
Battery noise floor for Li-Ion & LiFePo4?
Thanks for posting those measurements, it helps quantify the issue. In a previous discussion with Vinnie, iirc he said the LiFePo4 batteries had about 10 to 15 dB less broadband noise than SLA. If so, and based on the -96dB shown in your tests for the SLA, it would suggest a noise floor in the -110dB range for the LiFePo4...
It would be great if you could measure a LiFePo4 batt-pack under the same circumstances as the SLA in your tests, and see if the LiFePo4 is less susceptible to picking up line noise.
Thanks for posting those measurements, it helps quantify the issue. In a previous discussion with Vinnie, iirc he said the LiFePo4 batteries had about 10 to 15 dB less broadband noise than SLA. If so, and based on the -96dB shown in your tests for the SLA, it would suggest a noise floor in the -110dB range for the LiFePo4...
It would be great if you could measure a LiFePo4 batt-pack under the same circumstances as the SLA in your tests, and see if the LiFePo4 is less susceptible to picking up line noise.
Battery Impedance
No. I don't have the data, nor do I have the equipment to measure noise or impedance that low. I kluged together my test fixture, swept through my measurements, and decided on batteries. And considering I'm only powering a single LME49720HA op-amp with plenty of decoupling right next to the power leads, the batteries' Z-out in the low milliohm range was more than adequate.
The results are just what I designed them to be: The supply (batteries) and preamp output noise is below what I can measure (with my humble setup) and below what I can hear. It's also below what I've seen from several commercial preamps including an Adcom, a Threshold FET, and a couple tube units (my equipment can pickup the noise from the commercial gear).
I brought my DC preamp to Burning Amp 2011. We listened to it for a while and nobody complained ; )
I still do not think I can "hear" the batteries in my preamp, however.
No. I don't have the data, nor do I have the equipment to measure noise or impedance that low. I kluged together my test fixture, swept through my measurements, and decided on batteries. And considering I'm only powering a single LME49720HA op-amp with plenty of decoupling right next to the power leads, the batteries' Z-out in the low milliohm range was more than adequate.
The results are just what I designed them to be: The supply (batteries) and preamp output noise is below what I can measure (with my humble setup) and below what I can hear. It's also below what I've seen from several commercial preamps including an Adcom, a Threshold FET, and a couple tube units (my equipment can pickup the noise from the commercial gear).
I brought my DC preamp to Burning Amp 2011. We listened to it for a while and nobody complained ; )
I still do not think I can "hear" the batteries in my preamp, however.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.