Well, it's tempting to try two of those ebay boards in post #3, LT1083 Adjustable Regulated Power Supply Module DIY Kit Be | eBay
The IC alone costs twice as much locally as the ebay kit, hope it's not fake.
AndrewT, I don't see why it's so complicated, what can go wrong by trying two of those ebay boards? I intend to use them with lm3886 BrianGT chipamp.com boards.
The IC alone costs twice as much locally as the ebay kit, hope it's not fake.
AndrewT, I don't see why it's so complicated, what can go wrong by trying two of those ebay boards? I intend to use them with lm3886 BrianGT chipamp.com boards.
That's the problem: If you don't know how to design a well behaved power amplifier and don't know how to design a well behaved power regulator, then how can you possibly know how to get the combination to behave properly?
I don't know how to get the combination to behave properly, but the designing of the power amp and the power supply is already done.
I have found some tips for lm317, would some of this apply to other IC's?
Simple Voltage Regulators Part 2: Output Impedance
Using 3-pin regulators off-piste: part 3
And tomchr's neuromancer pages have good info on the amp.
Anyway, I don't think I'll bother with the regulated supply for now, I doubt I'll hear much difference, with such a small difference in THD, (0.003% vs 0.0006 - 0.003%). And the cheap ebay regulators are likely not as good as tomchr's regulated supply used for the tests, so less of a difference, or possibly even worse than unregulated could be expected.
I have found some tips for lm317, would some of this apply to other IC's?
Simple Voltage Regulators Part 2: Output Impedance
Using 3-pin regulators off-piste: part 3
And tomchr's neuromancer pages have good info on the amp.
Anyway, I don't think I'll bother with the regulated supply for now, I doubt I'll hear much difference, with such a small difference in THD, (0.003% vs 0.0006 - 0.003%). And the cheap ebay regulators are likely not as good as tomchr's regulated supply used for the tests, so less of a difference, or possibly even worse than unregulated could be expected.
I dunno... Building a power supply or a power amp are remarkably similar. A power amp can certainly be seen as a power supply regulator with a modulation input and wide regulation bandwidth. I think the main difference is that there are more plug-n-play solutions (aka chip amps) available in the amplifier space than in the supply regulator space. Besides, the OP did ask about regulators.
Tom
The THD may not be audible, but I bet the IMD products resulting from the supply interaction are.
Tom
I used the LT1083 boards with my LM1875 amps and they worked well. I increased the output capacitance to 4700uF/35VDC. Sounded very good to me with much lower noisefloor on the Chinese ampboards.
This is a very similar design to the old Audio Amateur LT1085/LT1033 KG-5 dual regulators. The LT1083 is the higher current version of the LT1085CT. The protection diodes are installed as the old design suggested.
Might try these with my LM3875 amps at some point to see if there is any difference from the standard supply.
This is a very similar design to the old Audio Amateur LT1085/LT1033 KG-5 dual regulators. The LT1083 is the higher current version of the LT1085CT. The protection diodes are installed as the old design suggested.
Might try these with my LM3875 amps at some point to see if there is any difference from the standard supply.
Tom, I wouldn't bet against you, but would you make that bet even with the LT1083 ebay boards?
I assume the regulated power supply you used is close to perfect?
I assume the regulated power supply you used is close to perfect?
I looked at the eBay module (linked to in Post #3). It uses an LT1083 in a package marked as obsolete by Linear Tech. It also uses a very small heat sink and no thermal pad (or goop) between the LT1083 and the heat sink. So it's off to a rough start. Including shipping, the boards will set you back about $6, so you're not out much if you want to try it out. I will point out that most LT chips cost more than $6 at Digikey and the TO-247 package is not a cheap package. Let's just say I'm a bit skeptical of the eBay module. I'd certainly load test it with a power resistor before putting it into service as the supply regulator for an amp.
That said, the LT1083 is a nice regulator. The data sheet shows 40 dB ripple rejection at 20 kHz, which is very nice. 7.5 A output current. Nice. Perfect for an LM3886 amp. I'm confident you'd be able to hit the data sheet performance of the LM3886 if you use the LT1083 as the supply regulator and use a good layout.
Reasonably close. The HP/Agilent/Keysight E3632A I was able to borrow is a pretty nice supply. 0.01 % load regulation according to its data sheet.
Tom
If anyone is wondering, I have about 10 of the LT1083 boards for projects. They typically have used LT1083 tested devices that work. I have them in projects at home without any failures to date after more than 1 year of use. YMMV.
Those are good findings, actually. I've made the same observations many times when I was designing ultra-low noise voltage regulators for use in precision timing products. Some newer regulators are marketed as "ceramic stable", i.e. stable with a low-ESR cap on the output, for this reason.
I'm pretty sure the "Eric Dietz" referred to in the last link is actually Errol Dietz. He used to be my skip-level manager. He's also worked for Bob Pease back in the day. Good guy.
Tom
I have been using them with the included heatsinks without heat buildup problems with the TO247 case. The latest batch I bought had a half height heatsink compared to the earlier units. Have not built them yet.
The pc board lead spacing for the 10 amp rectifiers does allow for 8 amp MUR860 diodes to be installed if you bend the leads appropriately.
The pc board lead spacing for the 10 amp rectifiers does allow for 8 amp MUR860 diodes to be installed if you bend the leads appropriately.
I've got a heap of quality heat sinks and thermal paste so that was the only modification I was going to make, aside from using a much larger cap.
Is testing them simply a matter of sending a few amps through some resistors and looking for ripple/voltage drops/excess heat?
Did you use the LM1875 in single-supply mode, then? I was understanding that the LT1083 was useful only as a positive-voltage regulator.
Used them as a bipolar supply with the two regulators stacked. The transformer has dual secondaries, dual rectifier bridges and caps. Worked well in that configuration with the LM1875's.
So is it OK to connect the Minus side of one regulator output to the Plus side of the second regulator in order to use them in a bipolar fashion?
As long as the connection between the two regulators becomes the ground connection and not the - output of the bottom regulator, it works correctly. Everything has to be floating for a load on the amp.
The first schematic on the webpage above is basically the same design as the LT1083 version. The LT1083 is a higher current capable device with very low output impedance.
OK, cool!! I just ordered two of the LT1083 regulator boards to use with my Elliott-powered PA speaker. Seems like a real bargain at $16 shipped for both; although I'll probably sub some 10,000uF caps that I have (if they'll fit).
- Status
- Not open for further replies.