Rechargeable Battery Supply for MC Step Up

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I’ve just finished a Boozhound step up and currently have it powered by 2x12V standard batteries with a simple switch. I plan to add another stack of 2x12V for increased playtime.

However, I’d like to be able to swap them out (or recharge them) before they get too depleted.

I’m looking for examples of power supplies (24V via 2x12Vx2 ideally to avoid chassis reconfiguration) that use rechargeable batteries, which switch to charging mode / mains power only when the battery is low. Maybe with LED indicator?

Should I just use regular lead acid batteries? Lithium ion?

I suppose it would have to switch from series to parallel to initiate charging from a 12V wall wart?

I’m assuming seamless charging to battery switching while listening is not advised?

An example would be the Ray Samuels Nighthawk ( Ray Samuels F-117 Nighthawk MC/MM Phono Preamplifier A real game changer in phonostage preamplification. Review By Wayne Donnelly )

Most out of the box battery circuits it seems are designed for non-interruption in the event of power failure.

And most diy examples I’ve seen here from searching are related to portable boombox type implementations.

Can anyone point me in the right direction? Has anyone designed something like this?

Is there something out there that’s drop-in that will provide minimal headache? That would be ideal. I’m short on DIY time these days unfortunately.
 
Member
Joined 2008
Paid Member
I’ve been through exactly what you are describing. I built the Boozehound standard MM phono stage several years back and went through many stages of trying to resolve a quiet power supply. I did not implement a system that automatically charges, but it is rechargeable and monitor-able. I also recently went through replacing the rechargeable supply in my ZYX CPP-1 in the same manner.

I had what I thought was a quiet laptop supply for a while but it crapped out. Ultimately I ended up with a couple 24V NiMH supplies from Batteryspace.com. I have no affiliation with them, they just had a good selection of battery supplies and chargers at reasonable prices.

I’ve attached a pic of a supply as they came to me as well as few pics of my solution.

Ultimately, I felt the 24V supply was over driving the circuit as the charger was bringing it up to about 26V on full charge and I was hearing distortion. I was plugging the supply into the 5.5X2.1 jack on the back of my case that I installed for the laptop supply. I wanted to integrate the supply as you mentioned and control charging as well as monitor the state of charge without having to use a multimeter and etc. I sort of used the ZYX as a model.

I opened up the 24V supply and removed two cells to bring it’s full charge capacity down to about 22V. The charger I bought still works with it as it can be used with anything from about 19-24V. Even the thermistor monitoring circuit still works properly. I simply had to bridge the last two cells on the altered pack with some wire and reinsulate the pack by completely rewrapping it with electrical tape. You can just buy a 22V pack to start with if you like.

I also used a multi switch system that has a center off 2 position switch to select between sending voltage directly to the battery for charge or using an external power supply to directly power the circuit. The battery charging supply is not appropriate to power the circuit. A secondary single pole switch controls whether the battery is connected to the circuit. A single LED indicates when power is reaching the circuit. I also integrated a momentary switch that activates a voltage meter I sourced off eBay so I can simply push a button to monitor the supplies voltage level. The voltage meters circuit was too noisy to be left running all the time.

I came up with a solution to use the voltage meters (since I got 3 for like $10) on the ZYX as well without modifying the unit. I machined a removable pod that plugs right into the power inlet on the ZYX. When the charge switch is activated it displays the battery supply voltage level. When it requires a charge I simply remove the pod and plug in the charger I bought for the 10.8V NiMH supply I installed in the ZYX (sourced from the same supplier).

I find these supplies are dead quiet and have over twice the capacity (in the case of the ZYX) of the original NiCD packs that were used.
 

Attachments

  • 2B2A586B-D6E4-4343-95EA-C132A2E85868.jpeg
    2B2A586B-D6E4-4343-95EA-C132A2E85868.jpeg
    687.8 KB · Views: 314
  • B23942E6-60B5-43D6-8AF4-93A36955C232.jpeg
    B23942E6-60B5-43D6-8AF4-93A36955C232.jpeg
    570.8 KB · Views: 304
  • E19E00D0-57C8-4DB4-B8DA-6B86872DB4A0.jpg
    E19E00D0-57C8-4DB4-B8DA-6B86872DB4A0.jpg
    562.7 KB · Views: 291
  • F7A81D76-BB79-497F-ABAA-AE0C0076518A.jpg
    F7A81D76-BB79-497F-ABAA-AE0C0076518A.jpg
    977.8 KB · Views: 292
  • 12134E0A-15BE-41BD-AA88-85C2EBEC168E.jpeg
    12134E0A-15BE-41BD-AA88-85C2EBEC168E.jpeg
    591 KB · Views: 284
Last edited:
Wow! Beautiful work!

I’ve been looking at Rod Elliott’s design here:

Automatic Charger for Battery Operated Hi-Fi Preamps

Ideally I’d prefer to have the charging seamless and not require removal. The likelihood I will be able to make such a neatly machined pluggable charging pack as you did is low.

But I will definitely take a closer look at you implementation, and get back to you with a more fleshed out response.

Thanks!
 
Last edited:
Member
Joined 2008
Paid Member
Wow! Beautiful work!

Thanks. Took me a while but I like to build things.

I’ve been looking at Rod Elliott’s design here:

Automatic Charger for Battery Operated Hi-Fi Preamps

I don’t see why you couldn’t use Rod’s article to try and build an automated charging system if you have the time and skill. I’m not sure if your Boozehound MC needs 12 or 24V, but he does outline building a +/-12V rechargeable supply. I’ve been to his site many times, but most times the projects like the one you pointed to seem a little over my head. I don’t have the confidence to fly solo and try to build a circuit from scratch with a diagram and he specifically says there is no support for the design in the article.

Ideally I’d prefer to have the charging seamless and not require removal. The likelihood I will be able to make such a neatly machined pluggable charging pack as you did is low.

As long as you have a way to connect and disconnect a charging system externally through switches and/or laptop style DC port there’s no reason not to mount the rechargeable pack in the case and leave it there regardless as to whether the charging system is automated.

The pod you refer to as a “charging pack” does not charge the battery or do anything related to charging. It is a removable voltage meter to monitor the charge level of the internal NiMH rechargeable power supply. It uses the same laptop style 5.5X2.1 power port on the front of the ZYX that my charger would have to be connected to in its place to charge the internal supply once I note that the level of the supply has fallen below the desired nominal voltage with the meter.
 
I also have very positive listening experiences when comparing a mains connected power supply versus a battery supply.
However a battery has a relatively high output impedance, so it could even be improved.
That's why I took a different road with very positive results.

I started with a SMPS wall supply delivering a DC voltage.
Two switches SW connect this supply to a large electrolytic cap C1 for 45msec, with 10 Hz switching frequency.
When the switches are open, a second set of switches SW1 transfer the load from the first cap to a second cap C2, interleaved and also during a period of 45 msec at 10 Hz.
At last an ultra low noise LDO Voltage regulator provides the wanted DC voltage.

In effect, there is never a connection between mains and the output voltage.
The caps acting as batteries provide the power for the equipment to be supplied.
The switches in my case are mosfets.
I have several versions working permanently for almost 4 years, equaling ca. 1Giga switches per supply so far, without a flaw.
I have called it a VAS, for Virtual Accu Supply. Maybe it's a bit complex, but it's a hobby after all.


Hans
 

Attachments

  • VAS.jpg
    VAS.jpg
    48.1 KB · Views: 128
However a battery has a relatively high output impedance, so it could even be improved.

The impedance isn't low by Jung regulator standards (micro-Ohms) but is in the hundreds of milli-Ohms for a lithium battery, in the tens of ohms for an alkaline battery.

Well advised to reduce the impedance above 100kHz by using a 100nF film cap, and bypassing locally. Why above 100kHz -- a strong signal from an AM broadcast station can modulate the power supply rails.

Battery impedance is a big topic with EV vehicles being all the rage!
 
Member
Joined 2011
Paid Member
When the switches are open, a second set of switches SW1 transfer the load from the first cap to a second cap C2, interleaved and also during a period of 45 msec at 10 Hz.

At last an ultra low noise LDO Voltage regulator provides the wanted DC voltage.

In effect, there is never a connection between mains and the output voltage.

Very, very nice! Congratulations.
 
Remember years ago Testings were published on battery power delivery.
Amazingly Nicads proved to be 'clean' beyond the abilities of the Lab equipment to measure . Far and away the superior Battery source choice.
Lotsa old Power tools used Nicads.. IF you don't have any of your own.
Beyond that I've also tried 3s lipos, Nimh round cells and lead acid/agm Moto battery even a silly simple Toshiba Laptop SMS brick.
Which I use currently.. as quite frankly I cannot hear ANY difference when using it.
Being analy fastidious is one thing ....but one's hearing AND Audio gear quality plays a Much more important part in all of this ;)
 
Last edited:
I also have very positive listening experiences when comparing a mains connected power supply versus a battery supply.
However a battery has a relatively high output impedance, so it could even be improved.
That's why I took a different road with very positive results.

I started with a SMPS wall supply delivering a DC voltage.
Two switches SW connect this supply to a large electrolytic cap C1 for 45msec, with 10 Hz switching frequency.
When the switches are open, a second set of switches SW1 transfer the load from the first cap to a second cap C2, interleaved and also during a period of 45 msec at 10 Hz.
At last an ultra low noise LDO Voltage regulator provides the wanted DC voltage.

In effect, there is never a connection between mains and the output voltage.
The caps acting as batteries provide the power for the equipment to be supplied.
The switches in my case are mosfets.
I have several versions working permanently for almost 4 years, equaling ca. 1Giga switches per supply so far, without a flaw.
I have called it a VAS, for Virtual Accu Supply. Maybe it's a bit complex, but it's a hobby after all.

Hans



Could you possibly explain in more detail this implementation, for the less capable among us (myself included).

It sounds very interesting.
 
Could you possibly explain in more detail this implementation, for the less capable among us (myself included).

It sounds very interesting.

Here it is in detail. Switching is performed by means of the Mosfets T1/T6 and T2/T7, steered by opto isolators with 10 Hz interleaved 45 msec signals Phase1 and Phase2.
The Caps to be charged are C1 and C2.
R11/R34 are used at start up to prevent large inrush currents. When the voltage on C2 has passed a border, U8 caused a relay to place R14/R15 in parallel.
To give you a better idea, the PCB is also shown.
Values of several components depend on the input and output voltages and the amount of current that has to be supplied.

Hans
 

Attachments

  • Untitled-1.jpg
    Untitled-1.jpg
    451.3 KB · Views: 132
  • Untitled-2.jpg
    Untitled-2.jpg
    388.6 KB · Views: 129
  • Untitled-3.jpg
    Untitled-3.jpg
    574.9 KB · Views: 121
Member
Joined 2007
Paid Member
Hi ... just briefly ...

If you use a couple of 12 VDC 7AH sealed lead-acid batteries they have an impedance around 10 mohms. Very few active circuits do better than this and they also have an extremely low noise floor (my - 153 dB noise floor measuring ADC cannot detect any noise contribution from such a battery even when discharging at 1A).

Such a battery can be float charged at 13.3 - 13.8 VDC and this may be done with e.g. an LM317T and LM337T.

If you want to disconnect the charger from the mains this can be done e.g. with a relay controlled by an LM555 timer - set for example to three hours so that serious listening can be done for three hours and then the relay automatically re-connects the charging circuitry.

Simple and outstanding sound IMHO.

Cheers,

Jesper

A P.S.: I prefer Panasonic SLA batteries.
 
Member
Joined 2011
Paid Member
Thank you very much for posting your schematics, Hans! May I inquire

why not install a large bypass capacitor between Node X and quiet-ground?

there seems to be a pathway (red) allowing input noise to make its way to the output, no matter which way the switches are connected; is it silly to worry about this pathway?​
 

Attachments

  • blobby.png
    blobby.png
    316.9 KB · Views: 125
Thank you very much for posting your schematics, Hans! May I inquire

why not install a large bypass capacitor between Node X and quiet-ground?

there seems to be a pathway (red) allowing input noise to make its way to the output, no matter which way the switches are connected; is it silly to worry about this pathway?


Hi Mark,

Resistors R9,21,22,25,27 force Sig$74 to stay Below V+ and Sig&81 above V-.
Any noise coming in at Node X from the input, will appear in Common Mode on Sig$74 and Sig&81.
And you are right, a bypass cap between node X and some quit ground could be beneficial, good point.

Hans
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.