Voltage Change Question
If I change the voltage of power supply from 24V to 20V, do I need to change parameters in Pearl circuit? Because I use shunt regulator instead of linear regulator. Voltage drop of my shunt regulator is very high. This circuit cannot produce 24V.
If I change the voltage of power supply from 24V to 20V, do I need to change parameters in Pearl circuit? Because I use shunt regulator instead of linear regulator. Voltage drop of my shunt regulator is very high. This circuit cannot produce 24V.
Last edited:
Shhhh, don't tell anyone you're doing that. You will need to change the LM3xxx's so that there is at least 2.5V overhead.
Correct me if I am wrong, but the relatively high voltage allows enormous and effortless overhead margin.
OTOH, I run the "Pearl Minnie" with one of Jan Didden's very quiet SMPS regulators -- you can use a pair of 18650's to run it for +/- 15VDC/
Hi all,
I want to thank everybody who contributed to this thread. Because reading it was a lot of fun en helped me to get my own Pearl 2 working the first time. Almost.
A few things may be worth mentioning and may help future builders.
I want to thank everybody who contributed to this thread. Because reading it was a lot of fun en helped me to get my own Pearl 2 working the first time. Almost.
A few things may be worth mentioning and may help future builders.
- I built it in a single enclosure. Because of lack of shelf space. And because I wanted to use the same enclosure that I got for my streaming DAC. It is giving me some noise when I up the volume. But at listening volume it is quiet, also when I put my ear right at the speaker. The noise I get is louder than de MM phono pre in my integrated amplifier with the volume knob at the same place. But it also has more gain. Anyway, not a problem for me at this point.
- The case can fit both boards next to each other, but I stacked them atop each other, inspired by seeing some in this thread use a 1U case. This way there is some space between the boards and the transformer. And it also means that I could make really short wiring for the input.
- Knowing that the ZVP3310 fail quite a lot I ordered 6 of them and put sockets in their place so I could swap them easily if needed.
- When I ordered at Mouser, they were out of ZTX450, so I used ZTX451 transistors. Seems to work fine.
- Beacuse I have both boards stacked, I first wired and tested just the right board. After that one worked, I installed the left board atop of it.
- The left board also worked, so I had a working Pearl 2. Nice!
- Only during startup I did get a few seconds of rumble in the left channel, nothing in the right channel. This was annoing, and something I told myslf I should look into. But then again, it worked fine. Just fire up the Pearl and after 10 seconds switch the amplifier to the channel and everything is fine.
- After a week though, the left channel stopped working. Completely silent. Luckily the left channel is the top board, which made trouble shooting easier.
- I found the problem pretty quick. The emitter of Q1 was 0V, which meant that Q3 and Q6-Q9 were everywhere at 0V. Luckily I also had 2 spare ZTX451 and could easily replace Q1.
- But why did Q1 fail? Looking and feeling around it seemed that C6 was feeling a little loose. And indeed, on the ground side I had a cold solder joint. After reflowing the I got voltages as expected and it is working again like a charm. And, without the rumbling sound in the left channel during startup. Happy days indeed.
Very nice @ bramvandij ! I'm thinking you could probably improve your noise situation further by installing a metal divider in the chassis between the PSU side and the amp side. Seems you have room for that if you move the isolation birdge a bit. You could probably also install the whole psu inside a smaller box that fit inside your chassi. That could potentially take things up even a notch further.
@bramvandijk -- the regulator IC's should be secured (screwed) onto the heatsinks -- with thermal grease and an insulating washer.
Has anyone replaced the 7824/7924 regulators with Belleson or Sparkos units? I'll be ordering one for another project, and I'm tempted to go for it on the Pearl 2. I also looked at a LT3045/3094 solution, but it won't go up to 24V, so they're out.
My PSU is done and tested, and the chassis is partially built. I'll get to PCB's in the coming weeks. I'm going to be making mine for use with MC cart primarily, but it will have a switch for 45/55/65dB gain.
My PSU is done and tested, and the chassis is partially built. I'll get to PCB's in the coming weeks. I'm going to be making mine for use with MC cart primarily, but it will have a switch for 45/55/65dB gain.
Attachments
The Belleson regulator doesn't win any prizes for noise. (See attached chart, noise should be V/RtHz) In fact, if you dig way back in this thread I measured the Pearl II noise with the conventional regulators and the Jung-Didden regulator and couldn't detect any difference.
If you can't find the fixed voltage regulators, you can use LM317/LM337 with 10uF from the ADJ pin to ground. In fact, it's much lower noise than the 7824/2924. You will have to make some adjustments to the mounting as the pinouts differ.
Attachments
Thanks Mark and Jack! I'll use the 7824/7924's I already have in the box of Pearl 2 parts.
@jackinnj - have you run similar tests on the Sparkos regulators? I don't know how similar or different they are compared to the Belleson. They seem conceptually similar, but I don't know the details.
@jackinnj - have you run similar tests on the Sparkos regulators? I don't know how similar or different they are compared to the Belleson. They seem conceptually similar, but I don't know the details.
I’m thinking of a single positive regulator for a line stage I’m working on. I’ll build the Pearl 2 to spec.
I purchased the Belleson to compare "blindly" to the regulators in the Linear Audio article. The Sparkos weren't on the horizon at that time. If they send a pair, I can re-run all the tests (PSRR, Noise, and Output Impedance).
I can say with ontological certitude that Elvee's noise reduction circuits are an inexpensive high quality addition to the pantheon of high quality regulators.
Coming along nicely. Just finished the PSU as the PCB's and FET's arrived from Pass. Unfortunately I did not see the "ugly psu cabinet" competition before I had ordered a rather nice box for my PSU
. Oh well, I guess i can hide it away anyway, and it's too small to stack components on anyway 
Populated the regulators and filter components. Voltages checked out. Started populating the rest of the resistor. Very nice boards by the way!Coming along nicely. Just finished the PSU as the PCB's and FET's arrived from Pass. Unfortunately I did not see the "ugly psu cabinet" competition before I had ordered a rather nice box for my PSU View attachment 1078594 View attachment 1078597 . Oh well, I guess i can hide it away anyway, and it's too small to stack components on anyway
Boards done, getting sound though, shows potential but some strange noises are present. That is probably related to a big oscillation at about 7.1 Mhz occuring on both channels. The first trace of the oscillation I see is at both source and drain of Q4. Nothing at the gate that I can see on my scope. So it originates in the "output stage" it seems. So we are getting positive feedback here somewhere, now to find out where and why. I have both c7 and c9 installed, no c15 and R15 is a jumper. Bias is fairly stable around 0 once warmed up a bit. Any Ideas what to try to get rid of the ocsillations? Searching this thread it seems possibly C7 may be the culprit so will try to take that one out.
Last edited:
And jupp, it was the c7's in my case as well. C7 gone and oscillations gone. Now this thing is playing music already,even when just connected like this.
Sure I am getting a little bit of hum but thats to expected. Now can't wait to get it boxed up. But also need to set gain for lomc. Now testing with MM Goldring 1042. Even that sounds scary good already. Thanks to those who had posted about the oscillation problem and solution!
Sure I am getting a little bit of hum but thats to expected. Now can't wait to get it boxed up. But also need to set gain for lomc. Now testing with MM Goldring 1042. Even that sounds scary good already. Thanks to those who had posted about the oscillation problem and solution!