Yes ! i was just staying ample ... very conservative 😱 if i will adopt 70V the trasformer could be quite small ...
Thanks a lot for the helpful advice. It seems to me that indeed the more sane choice is to go with a standard power supply based on a TL783 high voltage regulator
http://www.ti.com/lit/gpn/tl783
followed by some kind of passive filtering to cancel the last drop of ripple. You mention a 4-stage R-C filter ... i will look for schematic and calculator. Maybe i could use also some L ... but i really do not know about the more suitable values ... but i will use a standard supply for sure. That part is very very handy and convenient as well. Thanks a lot again.
Hi Gino,
I have just read briefly through the thread and would just like to add that batteries are indeed very low noise (in most cases close to unmeasurable) even when delivering quite some current/power. A couple of years ago I did some tests on SLA batteries (12V 7AH) and even when delivering 1A current the noise level was indiscernible to when not delivering any current. And it was at exactly the same level as when the AD-converter I used for the measurements had its inputs shorted (=lowest noise) ~ - 155 dB base noise level.
Also, the diyaudio member gerhard has made some noise measurements on other battery types and for just a reasonable one the noise level (to my memory) hovers around ~- 180 -> -200 dBs. Very low, indeed.
About series connecting SLA batteries it rarely is an issue if - for every ~27 VDC battery voltage - one also uses an over-voltage protection circuitry. This can e.g. be a TL431 - example circuities can be figure 18 & fig 19 in the attached datasheet cutout (from Motorola) depending on how much compensation current is needed.
In practice either of these shunt circuitries are adjusted to start drawing current when the voltage across the batteries to be protected reaches a voltage level defined by R1/R2 (can be a suitable value trimpot with center out). For SLA's a good standby voltage would be 13.5 VDC for a 12 VDC battery - and a maximum set to 13.8 VDC. The top left corner resistor could be e.g. 0.1 - 1 ohms again depending on shunt current draw (in a preamp case it likely would be less than 100 mAs if all batteries are equally new).
If you'd like to prevent the batteries from over-discharging - e.g. because you forgot to switch on the charge circuitry again - a 555 timer circuitry controling e.g. a power relay can be used to switch the charger back on after maybe 2-4 hours.
If you disconnect the charger while listening an LM317T in many cases is adequate to charge the batteries.
Remember that SLA batteries do not like higher temperatures (it shortens their life significantly) so any heat should be placed away from the batteries.
And then - and please note that this is entirely my own personal "view" so please feel welcome to disregard - if it were me I would try to find a preamp circuitry operating at a lower voltage. I don't personally use a preamp but I reckon that if you started a thread asking for preamp advice there would be many diyaudio members that could suggest very good solutions.
Hope this may help.
Cheers,
Jesper
I have just read briefly through the thread and would just like to add that batteries are indeed very low noise (in most cases close to unmeasurable) even when delivering quite some current/power. A couple of years ago I did some tests on SLA batteries (12V 7AH) and even when delivering 1A current the noise level was indiscernible to when not delivering any current. And it was at exactly the same level as when the AD-converter I used for the measurements had its inputs shorted (=lowest noise) ~ - 155 dB base noise level.
Also, the diyaudio member gerhard has made some noise measurements on other battery types and for just a reasonable one the noise level (to my memory) hovers around ~- 180 -> -200 dBs. Very low, indeed.
About series connecting SLA batteries it rarely is an issue if - for every ~27 VDC battery voltage - one also uses an over-voltage protection circuitry. This can e.g. be a TL431 - example circuities can be figure 18 & fig 19 in the attached datasheet cutout (from Motorola) depending on how much compensation current is needed.
In practice either of these shunt circuitries are adjusted to start drawing current when the voltage across the batteries to be protected reaches a voltage level defined by R1/R2 (can be a suitable value trimpot with center out). For SLA's a good standby voltage would be 13.5 VDC for a 12 VDC battery - and a maximum set to 13.8 VDC. The top left corner resistor could be e.g. 0.1 - 1 ohms again depending on shunt current draw (in a preamp case it likely would be less than 100 mAs if all batteries are equally new).
If you'd like to prevent the batteries from over-discharging - e.g. because you forgot to switch on the charge circuitry again - a 555 timer circuitry controling e.g. a power relay can be used to switch the charger back on after maybe 2-4 hours.
If you disconnect the charger while listening an LM317T in many cases is adequate to charge the batteries.
Remember that SLA batteries do not like higher temperatures (it shortens their life significantly) so any heat should be placed away from the batteries.
And then - and please note that this is entirely my own personal "view" so please feel welcome to disregard - if it were me I would try to find a preamp circuitry operating at a lower voltage. I don't personally use a preamp but I reckon that if you started a thread asking for preamp advice there would be many diyaudio members that could suggest very good solutions.
Hope this may help.
Cheers,
Jesper
Attachments
Last edited:
Hi Jesper ! as maybe you have understood i am trying to put together a decent preamp with very minimalistic topology. I do not know nothing about the PSRR of the circuit, but a very low noise PS is good to have for a line preamp. I am looking for something very low noise and very low distortion. And batteries seemed to be an option ... but i am feeling i have open the classic can full of worms ... not very easy to do rightly.
😱😱😱 do you have any link to pages describing that AD converter ? impressive low noise indeed. very very very interesting
i will try to trace that thread for sure. I remember a paper on batteries noise let me see ... here it is
https://tf.nist.gov/general/pdf/1133.pdf
(conclusions on page 373)
Dear Jesper it helps a lot. Actually i think i have made up my mind. I will evaluate the opportunity to use some rechargeable battery of the type used for electrical tools ... very cheap and with very handy recharger. wow ... they are not cheap at all but they have also higher voltage like 36V here
https://www.amazon.com/Replace-Lith...prefix=BLACK+DECKER+battery+36,aps,232&sr=8-5
i can optimized easily the circuit for the lower 36V voltage available with sim software (circuit current draw should be less than 50mA typically) The idea could be to left the battery on recharge when i do not listen the preamp and switch the recharger off when i listen to music ... What do you think ? could it be worth checking the solution out ? 😕 Thanks sincerely again 🙂
Last edited:
Hi Gino,
The AD converter is the AD7760 from Analog devices. There's an evaluation board available which is what I use (powered by a battery). Do note though that the input is balanced and the input impedance is 1k per balance input.
Regarding using Li-ion batteries this may be quite different than using SLA batteries. Different chemistries and different ways to charge them. If you search the internet you may find graphs showing how to charge Li-ion batteries.
I have, however, been using Li-ion batteries for many years now at a 4.05 VDC standby voltage per battery cell (that is pure Li-ion batteries NOT modified chemistries). It appears to work fine. But in my experience Li-ion batteries need over-voltage protection for each cell as they otherwise DO tend to get out of balance. Some cells may severely over-charge if not balanced.
I may also mention that SLA batteries in practice may have (much) lower impedances than series connected Li-ions. Such data may be found on the internet. I have good experience with Panasonic & Yuasa SLA batteries.
I do not particularly know about the battery unit you posted a picture of but when buying such battery units it often is unknown which brand & type the batteries are. I reckon it could also be Lithium polymer batteries which are different from Li-ions.
BTW the member iancanada has designed a LiFePO4 battery system based on A123systems cells. Again, however, an SLA 12V 7 AH battery has lower impedance and actually is quite simple and rugged to work with.
And then ... when it comes to sound quality I have not found anything that compares with batteries (essentially for all types of audio circuitries). A personal remark, I know 🙄
Good luck in your endeavors 😉
Jesper
The AD converter is the AD7760 from Analog devices. There's an evaluation board available which is what I use (powered by a battery). Do note though that the input is balanced and the input impedance is 1k per balance input.
Regarding using Li-ion batteries this may be quite different than using SLA batteries. Different chemistries and different ways to charge them. If you search the internet you may find graphs showing how to charge Li-ion batteries.
I have, however, been using Li-ion batteries for many years now at a 4.05 VDC standby voltage per battery cell (that is pure Li-ion batteries NOT modified chemistries). It appears to work fine. But in my experience Li-ion batteries need over-voltage protection for each cell as they otherwise DO tend to get out of balance. Some cells may severely over-charge if not balanced.
I may also mention that SLA batteries in practice may have (much) lower impedances than series connected Li-ions. Such data may be found on the internet. I have good experience with Panasonic & Yuasa SLA batteries.
I do not particularly know about the battery unit you posted a picture of but when buying such battery units it often is unknown which brand & type the batteries are. I reckon it could also be Lithium polymer batteries which are different from Li-ions.
BTW the member iancanada has designed a LiFePO4 battery system based on A123systems cells. Again, however, an SLA 12V 7 AH battery has lower impedance and actually is quite simple and rugged to work with.
And then ... when it comes to sound quality I have not found anything that compares with batteries (essentially for all types of audio circuitries). A personal remark, I know 🙄
Good luck in your endeavors 😉
Jesper
Do you even have an idea how much ripple is acceptable? (Remember the preamp's designer also thought about this-- not commercially profitable to design a preamp which needs a power supply more expensive than the audio-bits.)
Just with junk-box parts, I could get sub-microvolt ripple from raw AC. Yes, it looks like a fine sine wave but this is an extreme blow-up. (One advantage of multiple R-C is that the higher more-audible harmonics go WAY down.)
Attachments
Hi Jesper ... the fact that the resulting noise is -155dB down must mean that or the evaluation board has an unbelievable PSRR or the battery is really silent 🙄
A very valuable evidence for battery low noise indeed.
Very interesting thank you. However in the meantime i have found batteries used for small electrical tools providing 40V and 2Ah ... my circuit will need no more that 100mA
https://www.amazon.com/Greenworks-40V-Lithium-Battery-29462/dp/B00AW72XFQ
could i lower the impedance by placing a big caps after the battery ? maybe it will work fine.
thank you so much again. The more i think about batteries and the more i am scared ... i understand that some types are carried inside metal boxes because they can catch fire when pierced ... moreover i see very few people using it to power their audio equipment ... i do not know really 😕😕😕
Hi ! no idea at all. It is for a line preamp. Actually two ... one will be based on the Nelson Pass design Bride of Zen ... the other one will be taken by an old schematic from 70s ...
Both will use around 60V supply ...
i guess it would be more than enough for me really ... there will be the PSRR of the amplification circuit (know nothing about it anyway)
But your result is really excellent
Thanks ! very very nice. Which SW have you used ? i love the output. I would like to buy one copy. Is it expensive ?
I guess this solves completely my problem ... ok i can add a high voltage TL783 regulator at the beginnig ... but this is really impressive.
The circuit will be class A so current draw (no more than 100mA) will be constant.
i really fell in love with this SW i am serious
what is this ? can it simulate also power supplies with active parts inside ?
is it very expensive ? and models ... are expensive ?
what is this ? can it simulate also power supplies with active parts inside ?
is it very expensive ? and models ... are expensive ?
@Gino:
Please note that it is the "base noise level" of the AD converter that is at ~ - 155 dB (relative to appr. 3.4 Volt peak). And it is at - to my memory - a 64 kHz sampling frequency which means that the samples have been "averaged". The actual sampling frequency of the AD converter is 2.5 MHz and at this sampling frequency (no averaging) the base noise level is ~- 110 dB (again to my memory). However, when supplying it from a battery the - 155 dB noise level is virtually straight down to 0 Hz frequency which I have not seen when supplying the AD converter from a more normal PSU.
Regarding using a battery or not I might suggest a simulation software called LTSpice. It is free for download and there's a thread here on diyaudio dealing with getting from "beginner" to "advanced". Although one still has to have a good electronics knowledge in my experience it has helped clarify many questions I have had in relation to electronics and various circuitries. The thread is here:
Installing and using LTspice IV (now including LTXVII). From beginner to advanced.
FYI I have simulated the circuitry you posted above with respect to its frequency characteristics (in LTspice) - it is attached.
In case you decide to use batteries if I were a not too experienced battery user I would go with the SLA types because they IMHO are more rugged and less difficult to work with. SLA batteries are similar to car batteries - but they are also sealed so that the battery fluid may not accidentally be spilled.
Good luck in your endeavors ...
Jesper
Please note that it is the "base noise level" of the AD converter that is at ~ - 155 dB (relative to appr. 3.4 Volt peak). And it is at - to my memory - a 64 kHz sampling frequency which means that the samples have been "averaged". The actual sampling frequency of the AD converter is 2.5 MHz and at this sampling frequency (no averaging) the base noise level is ~- 110 dB (again to my memory). However, when supplying it from a battery the - 155 dB noise level is virtually straight down to 0 Hz frequency which I have not seen when supplying the AD converter from a more normal PSU.
Regarding using a battery or not I might suggest a simulation software called LTSpice. It is free for download and there's a thread here on diyaudio dealing with getting from "beginner" to "advanced". Although one still has to have a good electronics knowledge in my experience it has helped clarify many questions I have had in relation to electronics and various circuitries. The thread is here:
Installing and using LTspice IV (now including LTXVII). From beginner to advanced.
FYI I have simulated the circuitry you posted above with respect to its frequency characteristics (in LTspice) - it is attached.
In case you decide to use batteries if I were a not too experienced battery user I would go with the SLA types because they IMHO are more rugged and less difficult to work with. SLA batteries are similar to car batteries - but they are also sealed so that the battery fluid may not accidentally be spilled.
Good luck in your endeavors ...
Jesper
Attachments
Hi ! yes 😉 i guess is PSU Designer II ? 🙄
i read "to help you with the design of simple linear (unregulated) mains power supplies"
i think it is very interesting for filters ... but usually for preamp i see regulated power supplies ... actually i think to use some bjt/darlington to smooth the voltage before the filters.
I will look for a very simple but suitable schematic to get around 60V ...
Thanks again
🙂Hi ! i think i understand ... however batteries win anyway. Even the best power supply has some residual ac ripple ... even if sometimes i see amazing graphs indeed
(Prism Audio Lyra 2 noise floor) i understand that Prism Audio make also instrumentation ... audio analyzer for professionals. And it shows ...
i am starting studying it ... thank you ! a great SW indeed
I run some simulations in the past but i lost the files ... i have to start again
Really great SW
Jesper, i think i have made up my mind ... batteries are too challenging for me
I understand they can be the 1st choice for measurement equipment ... but are too complicated.
I will start looking at regulated power supply ...
Thank you so much again for all your kind and very precious advice.
Kind regards, gino 😀
Last edited: