Single supply LM1875 on a double supply pcb

[iinself]

Hi,
I have an old receiver with a PSU +/- 58V. I want to use the +58V rail to build a LM1875 amp. I have some dual supply LM1875 pcbs from ebay. I was wondering if I can use the pcb to build the single supply LM1875 amp.

This is the pcb I have - 2pcs LM1875T LM675 TDA2030 TDA2030A Audio Power Amplifier PCB DIY | eBay

I got a bunch of 63V caps with the needed values as in the datasheet.

Thanks.

 

[Arty]

unlikely a good solution.
i would search for other alternatives, like a more suitable powersupply,
or a poweramp ic that can take advantage of that high supply voltage.
or modify the transformer if possible,
or trade the transformer,
or try a virtual ground.
you might get away with a single 5a regulator to create a virtual ground at one of the rails.
set the regulator to half the single rail voltage (that is 58/2 volts) add a series diode on both rails to get it even a bit lower.
then it will fit the lm1875 split suppl boards, but its still a hack solution.

 

[Arty]

this would do even better:

An externally hosted image should be here but it was not working when we last tested it.

Virtual Ground (regulated!) - and Rail Splitter Circuits!

here is the site with information on how to calculate for say.. +- 28 volts.
You would still need some diodes to drop the powersupply voltage a bit.

 

[iinself]

I considered LM350/337 solutions but took +58V and -58V as the input and that led to a lot of power dissipation in the LMs. But I think I can use the solution for just +58V and remove the other -58V cap, it is a voltage double circuit after the rectifier. This way I can split the 58V to +29 - -29 using the above circuit.

Thanks

 

[Arty]

should work well with the above circuit, it keeps the virtual ground quite close to center.
but remember, it may drift a bit with load. allso mains voltage can change a bit.
aiming for +-29 volt is a bit dangerous, it is very close to the absolute maximum rating of the lm1875.
i would use a pair of diodes to scrape off a volt or so from the rails. done before the splitter will reduce the load on the regulators a tiny bit, and allows for a safety margin so you do not pop the chips.

 

[iinself]

Studying the circuit, I wonder how it will be able to supply about 3A. Since the current is limited by R1/R2, if I drop it 1ohm then I will need to use 9 watts resistor. For 2 resistors it will be 18 watts of loss, don't you think it is a bit too much?

 

[Arty]

why would current be limited by R1 and R2 ?
"The LM336BZ-2.5V voltage reference compensates for the LM317's (+1.25V) internal reference and the LM337's (-1.25V) internal reference. So when the LM317/LM337 adjust pins are connected inside the R1/R2 voltage divider as shown, each voltage regulator output voltage becomes 1/2 of whatever the rail-to-rail voltage happens to be."

those only set the regulator's output voltage, if i'm not mistaking.

 

[iinself]

Yes you are right.

 

[Arty]

any updates ?

 

[iinself]

Googling around I found this simpler version of the circuit. The first picture here http://www.goldpt.com/virtual_ground_circuit.html

Since my 56V is coming from a decent PSU I will be use the simpler version. I am struggling to find the LM336, I may have to go with TL431 instead.

One issue I am having is I will need about 3A through each of the regulators, I am hoping I can use a transistor for high current as given in the LM317 datasheet.

 

[sofaspud]

A couple of things for your consideration:
Is 58V the loaded secondary VAC? If not, it may help to get a more accurate number for design calculations.
The voltage regulation can be in multiple stages to lower single-device power dissipation.

 

[iinself]

Did not notice that the first circuit in your link is the same as the one in link I posted.

Can I use a normal diode forward biased as the reference voltage, why do I need to use LM336 or other reference regulators?

 

[johnr66]

+/- 22.5 volts (45v) is as far as I'd take the LM1875 with an 8 ohm (rated) speaker. The varying impedance of a loudspeaker load can be an issue. With 4 ohm load, +/-18 volts (36v) is as far as I would go. Some would say 36v is too much!

 

[Arty]

supposedly if i'm not mistaken, the solution in the ciruit is used to make sure the regulators do not try to fight against eatch other, and keep the ground dead on center.
and you can parallel regulators.

 

[6L6]

iinself - You are going to spend more money making a regulator to use your existing transformer -- its cheaper to buy the right transformer and go from there...

And, a 317/337 reg won't have the current capability you want...

The LM1875 has fantastic Power Supply noise rejection, the PSU need be no more than a bridge and two largish capacitors.

If you are dying to make the PSU from your transformer, (not a particularly good solution) and want to make the above regulator ( which doesn't have the current you need) Here is a link to the LM336 2.5V -- LM336Z-2.5/NOPB Texas Instruments | Mouser

 

[Arty]

You are correct, would be better to get the right Xformer straight.
but, if the OP wishes to stick to thisone, then its no problem.
There are 5A capable regulators available, not sure on negative voltage regulators to be honest. Surely lm1875 can get away with a diode bridge and a pair of caps, but meanwhile it can benefit from a regulated supply.

 

[bentsnake]

,,,You are going to spend more money making a regulator to use your existing transformer -- its cheaper to buy the right transformer and go from there...

You speak rightly. Something of a best kept secret are RadioShack's power transformers:

25.2V CT 2.0A Heavy-Duty Chassis-Mount Transformer with Lead - Radioshack
RadioShack® 25.2V CT 450mA Std Chassis-Mount Xformer - Radioshack

At $15 for their 25.2VCT, 50VA size--presently on sale for $13--it's about the best deal in town. Also handy is their perfboard #2761396.

The shack is going through some financial turmoil right now, so parts might not be on the shelf in every store. But you can check local stock on the web, and orders over $25 ship free.

PS for those who don't know: watts = volts x amps = VA
.

 

[Arty]

"
watts = volts x amps = VA
"

eh.. simply:
NO.

 

[iinself]

iinself - You are going to spend more money making a regulator to use your existing transformer -- its cheaper to buy the right transformer and go from there...

And, a 317/337 reg won't have the current capability you want...

The LM1875 has fantastic Power Supply noise rejection, the PSU need be no more than a bridge and two largish capacitors.

If you are dying to make the PSU from your transformer, (not a particularly good solution) and want to make the above regulator ( which doesn't have the current you need) Here is a link to the LM336 2.5V -- LM336Z-2.5/NOPB Texas Instruments | Mouser

Actually you can make LM317 supply more current by using a diode and a transistor, it is there in the LM350 sheet. So the "current" idea is to reduce the 58V to about 48V and use the single supply circuit in the LM1875 datasheet.

 

[bentsnake]

"
watts = volts x amps = VA
"

eh.. simply:
NO.

To the novice: this helpful forum member's informative post is correct. Watts and VA are not equal in an engineering sense. You have to take into account a thing called "power factor."

However, for DIY audio purposes we can assume watts and VA to be essentially the same. We use quick rules of thumb like this because they're accurate enough for non-engineering purposes, and they get the job done without complicated math.

For those wishing to be strictly correct, here's a calculator that will make the watts <~> VA conversion precisely. No doubt the helpful member intended to supply such a link, but forgot:

Watts to VA conversion calculator

I, for one, wish to thank the helpful member for taking the time and trouble to share his expertise with the rest of us. It's posts like his that make these forums a valuable source of help for beginner and expert alike.
.