A little background first --
I got the boards and Q pack from Pass DIY eight or nine years ago and I finally got around to building the Pearl in 2006. It's been an excellent phono stage and I love listening wo my old vinyl through it. I use a little Creek 4330 integrated amp with a MM phono stage and the difference between the Creek board and the Pearl is pretty stunning really.
A few months ago I was getting crackling in one channel. I didn't use the Pearl much after that but recently I decided to see if I could fix it.
So I fired it up and wasn't getting anything from the channel that had been crackling. I started checking voltages and when I got to Q4 it looked like it had overheated and fried. That looked like the problem and would certainly explain the crackle so I ordered some new ZTX450s from DigiKey.
I replaced the original Q4, powered it up and started checking voltages again. The voltages were off (significantly) around Q4 and there was still no sound from the channel.
After reading quite a few threads here it seems like the problem may be the 2SK389 but I don't really know how to be sure. Anyone want to help direct me through the troubleshooting process to narrow it down?
Also, if it is the 2SK389 that looks like it's a problem since that part isn't in production any longer. If that's the problem what are my options for fixing the Pearl?
Thanks in advance for any help you all can lend, it's appreciated.
I got the boards and Q pack from Pass DIY eight or nine years ago and I finally got around to building the Pearl in 2006. It's been an excellent phono stage and I love listening wo my old vinyl through it. I use a little Creek 4330 integrated amp with a MM phono stage and the difference between the Creek board and the Pearl is pretty stunning really.
A few months ago I was getting crackling in one channel. I didn't use the Pearl much after that but recently I decided to see if I could fix it.
So I fired it up and wasn't getting anything from the channel that had been crackling. I started checking voltages and when I got to Q4 it looked like it had overheated and fried. That looked like the problem and would certainly explain the crackle so I ordered some new ZTX450s from DigiKey.
I replaced the original Q4, powered it up and started checking voltages again. The voltages were off (significantly) around Q4 and there was still no sound from the channel.
After reading quite a few threads here it seems like the problem may be the 2SK389 but I don't really know how to be sure. Anyone want to help direct me through the troubleshooting process to narrow it down?
Also, if it is the 2SK389 that looks like it's a problem since that part isn't in production any longer. If that's the problem what are my options for fixing the Pearl?
Thanks in advance for any help you all can lend, it's appreciated.
Just build the new Pearl coming out soon 
it's said the boards will be the same size, you you can just drop them
in to the chassis that you already have been using. viola! better Pearl.
(maybe check to make sure you don't have a problem with the power supply,
something caused a problem to take out that part.)
(double check all solder points - crackling equals cold joints?)
it's said the boards will be the same size, you you can just drop them
in to the chassis that you already have been using. viola! better Pearl.
(maybe check to make sure you don't have a problem with the power supply,
something caused a problem to take out that part.)
(double check all solder points - crackling equals cold joints?)
I don't know much about this, so forgive me but is that all there is to it? Just drop a pair of 2SK170BL in where the 2SK389 was? Seems like there are seven leads from the 389.
theres a group by goin on by spencer from HK. He sells matched quads of 2SK170BL which fit perfectly. Just swap BOTH 2SK389 to 2 pcs 2SK170 each!
cheers Uli
Seriously considering the Pearl 2 and I probably will do that. But I hate to leave the original one if it can be fixed easily, it's been a really nice piece of gear.
Just build the new Pearl coming out soon
it's said the boards will be the same size, you you can just drop them
in to the chassis that you already have been using. viola! better Pearl.
(maybe check to make sure you don't have a problem with the power supply,
something caused a problem to take out that part.)
(double check all solder points - crackling equals cold joints?)
The LSK389 is a direct replacement for the 2sk389. There are also a couple of vendors who sell these parts at a premium. I think I found some of the 2sk at $15 each. After that I jumped in on the LSK389 group buy and picked up some spares.
The only way I could figure out how to troubleshoot my pearl (with this same problem) was to replace the 389. And then it started making magic again.
The only way I could figure out how to troubleshoot my pearl (with this same problem) was to replace the 389. And then it started making magic again.
your options are a matched pair of 2SK170 singles (or the physically smaller replacement) or this from linear systems: http://www.linearsystems.com/datasheets/LSK389.pdf
mlloyd1
mlloyd1
Hey Windchill,
the 2SK389 is nothing but a pair of matched 2SK170. The seventh pin on the 389 is a so called substrate pin and not at all needed.
If the problem is the 389, consider the exchange an implement 2 2SK170 instead. Most propably the matched 2SK170 will even turn out to be closer in IDSS as the two parts of the 2SK389 were.
But now I remember a friend once telling me about problems with the Pearl. I don´t know if this is the case here but he told me about issues related to the 2SK389 being solved by simply using individual source resistors (R25 (value of 10R) for each half of the diff amp).
So you might want to check the 2SK389 first for integrity. In case you have to change it, there is a guy from Sweden I believe. He is here in the forum as well and he had a bunch of 2SK389 a while ago. Otherwise, take 2 matched 2SK170...
Cheers
Chris
the 2SK389 is nothing but a pair of matched 2SK170. The seventh pin on the 389 is a so called substrate pin and not at all needed.
If the problem is the 389, consider the exchange an implement 2 2SK170 instead. Most propably the matched 2SK170 will even turn out to be closer in IDSS as the two parts of the 2SK389 were.
But now I remember a friend once telling me about problems with the Pearl. I don´t know if this is the case here but he told me about issues related to the 2SK389 being solved by simply using individual source resistors (R25 (value of 10R) for each half of the diff amp).
So you might want to check the 2SK389 first for integrity. In case you have to change it, there is a guy from Sweden I believe. He is here in the forum as well and he had a bunch of 2SK389 a while ago. Otherwise, take 2 matched 2SK170...
Cheers
Chris
Thanks everyone. I decided to just replace the 2SK389s in both boards and ordered the 2SK170s from Spencer here on the forums. Once I get those in we'll take another look and see how it is.
Chris, how would I check the integrity of the 2SK389?
Thanks again guys, I learned a lot and I appreciate your time and help.
Chris, how would I check the integrity of the 2SK389?
Thanks again guys, I learned a lot and I appreciate your time and help.
Pearl Phono V1 PCB Grounding system - confused
Hi everyone, thought I would piggyback this thread with my query.
I am building v1 PCB's, followed the article construction notes, but when I fired up and checked voltages they were all over the shop.
Turns out the PCB has a large number of ground points that are "dead ends" not connected to a pcb ground plane -- not connected to anything in fact. C2, C2, C3, R12, etc etc etc.
(I got my pcb's from promethius but I assume they are as per the 2001 article. The PCB has a note to insert a single link from C15 negative terminal to a ground point but that only fixes C15.)
What is going on? There is no mention of this in the article. Am I supposed to run ten-plus separate wires from all these dead ends to a common ground point - which the article says is at the output RCA sockets?
I now suspect that R6 and R12 are open circuit (measured in circuit) - and possibly Q2 is damaged as it is between R6 and R12. Is there a simple test of Q2?
Thanks for any help,
Arg
Hi everyone, thought I would piggyback this thread with my query.
I am building v1 PCB's, followed the article construction notes, but when I fired up and checked voltages they were all over the shop.
Turns out the PCB has a large number of ground points that are "dead ends" not connected to a pcb ground plane -- not connected to anything in fact. C2, C2, C3, R12, etc etc etc.
(I got my pcb's from promethius but I assume they are as per the 2001 article. The PCB has a note to insert a single link from C15 negative terminal to a ground point but that only fixes C15.)
What is going on? There is no mention of this in the article. Am I supposed to run ten-plus separate wires from all these dead ends to a common ground point - which the article says is at the output RCA sockets?
I now suspect that R6 and R12 are open circuit (measured in circuit) - and possibly Q2 is damaged as it is between R6 and R12. Is there a simple test of Q2?
Thanks for any help,
Arg
Hi Arg,
I also build one with Promitheus boards (by the way, my boards are identical to the Pass-boards as they've been made with the same files); if you inserted the link, all grounds should be ok.
However, if they are not (as you write), then check the 2nd channel wether it shows the same.
No, C15 ground should be connected to C1, C2, C3 ground (and not to ground plane). The link should connect signal stage ground (connected to ground plane) to power supply ground.
If the grounds of C1, C2, C3 are not connected even without the link, I expect that you maybe got a defective pcb. That also happened to me, but I had a short from rail to ground. Was hard to find, and nearly impossible to see as it was a tiny copper hair line on the board.
Have fun, Hannes
EDIT:
Not without unsoldering the part. Generally these are sturdy parts and I would not think that it's damaged.
I also build one with Promitheus boards (by the way, my boards are identical to the Pass-boards as they've been made with the same files); if you inserted the link, all grounds should be ok.
However, if they are not (as you write), then check the 2nd channel wether it shows the same.
No, C15 ground should be connected to C1, C2, C3 ground (and not to ground plane). The link should connect signal stage ground (connected to ground plane) to power supply ground.
If the grounds of C1, C2, C3 are not connected even without the link, I expect that you maybe got a defective pcb. That also happened to me, but I had a short from rail to ground. Was hard to find, and nearly impossible to see as it was a tiny copper hair line on the board.
Have fun, Hannes
EDIT:
Not without unsoldering the part. Generally these are sturdy parts and I would not think that it's damaged.
Last edited:
@WindChill
I don't think Q4 was fried first as it passes only little current and has little voltage across it.
I think the 2SK389 blew and sort-of shorted drain to source. When that happens Q4 sees almost 14 V Base-to-emitter voltage which is nearly three times above its rating of 5V. Q4 is fried then too.
Think about replacing the 2SK389 with 2x 2SK170 (don't need to be matched at all), replace Q4 and if you are at it, replace also R14 (1k) as it likely had to run above its rated power dissipation (all 29V times 20-30mA~ 1 W for a part rated at half of it).
Have fun, Hannes
PS: when unsoldered, you can check IDSS (9V battery connected to small resistor (10-100R), resistor connected to drain, other battery pole to source, gate connected to source) of the 2SK389 to see wether I'm wrong and the part is ok.
EDIT: in the unlikely case you have 2SK369GR at hand, you could use a single one to replace the 2SK389.
I don't think Q4 was fried first as it passes only little current and has little voltage across it.
I think the 2SK389 blew and sort-of shorted drain to source. When that happens Q4 sees almost 14 V Base-to-emitter voltage which is nearly three times above its rating of 5V. Q4 is fried then too.
Think about replacing the 2SK389 with 2x 2SK170 (don't need to be matched at all), replace Q4 and if you are at it, replace also R14 (1k) as it likely had to run above its rated power dissipation (all 29V times 20-30mA~ 1 W for a part rated at half of it).
Have fun, Hannes
PS: when unsoldered, you can check IDSS (9V battery connected to small resistor (10-100R), resistor connected to drain, other battery pole to source, gate connected to source) of the 2SK389 to see wether I'm wrong and the part is ok.
EDIT: in the unlikely case you have 2SK369GR at hand, you could use a single one to replace the 2SK389.
Last edited:
It's been a little while but the 2SK170BLs from Spencer took over a month to arrive. I replaced the 2SK389BL double FETs, rebuilt each channel's power supplies so I could be sure they were solid. Each channel is getting 40.0V now.
My voltages still look off in the same area. I went through the voltages specified in the Pearl article in Figure 5 and for the most part they're reasonably close until Q4, Q5 (one reading) and Q6.
Here are my readings from the bottom section with the expected voltages from the schematic in parentheses :
24.5 (24.9)
23.9 (24.2)
Around Q5
11.5 (11.6)
10.8 (10.9)
14.0 (15)
28.4 (28.8)
Around Q4
16.4 (14)
17.3 (19.3)
16.4 (13.4)
I just checked R14 and it reads 1.01K.
Any advice is greatly appreciated.
Wayne notes around 40V, and the silk screening on the boards show DC in voltage is 36V-45V.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.