Hi All,
I am hoping someone out there can direct me in the right direction. My Theta DSPro DAC quit working one day. I dug into it and discovered that one of the PSU LM317T regulators on the 5v rail was not working. I replaced it and all the other regulators (UR2 thru UR5). I also went ahead and replaced all of the electrolytic caps on the PSU board. In addition, one of the rectifiers W02G (BR1) on the PSU+8/+5 logic board rail was suspect so I replaced it with a 2W02G, since I could not find a reliable source for another w02G rectifier.
I have attached pics of the PSU board. In testing the voltage outputs, I discovered that the +8V output going to the logic board was putting out 14V. To me this is way too high. The PSU logic board rail has two outputs: a +5V and a +8V tap (see IMG_2427.jpg) both going to the processor board.
I am unsure of what the +8V tap was putting out BEFORE the repairs. I also am unsure whether the PSU has ever been touched since its early to mid 90's manufacture date. It appeared that all onboard components were factory default. However I do know that it is now putting out just a tad over 14V (14.1 to be exact) at the +8V output tap.
My question is this, how do I get the 14V close to the spec 8V? Or, is the 14V the right "non load" voltage?
I suspect the answer lies in the LM317K realm somehow. I have found LM317 "calculators" online and have calculated the resistor values that should result in the correct voltage, but they didn't work... either the +8V or 5V were incorrect (frequently both were)
The reading of the LM317K voltages are as follows:
Vin=13.9V
Vadj=3.7V
Vout=4.99V
Readings at BR1 "+" pad is 13.9V
In playing around with different values of R1 (factory value: 75R), R2 (factory value: 715R), and R3 (factory value: 243R) I have been unable to get the +14V down without affecting the Vout voltage. Measurements at R1 are 13.9-14V (the same reading at the +8V output tap on the PSU board). It appears that the 13.99 - 14.1V output is what is coming out of the BR1's "+" pin which is what is being converted from the transformer? Hence, my understanding is that the 14V goes thru R1 and C5 and ends at the +8V output pad (which goes on to the logic/processor board). So another question I have is what does R1 do? Reduce current/voltage across the R1 terminals?
What am I missing?
I would greatly appreciate any guidance, as I am dead in the water at this point, and don't want to put it all together only to fry circuits of the logic/processor board!
I am hoping someone out there can direct me in the right direction. My Theta DSPro DAC quit working one day. I dug into it and discovered that one of the PSU LM317T regulators on the 5v rail was not working. I replaced it and all the other regulators (UR2 thru UR5). I also went ahead and replaced all of the electrolytic caps on the PSU board. In addition, one of the rectifiers W02G (BR1) on the PSU+8/+5 logic board rail was suspect so I replaced it with a 2W02G, since I could not find a reliable source for another w02G rectifier.
I have attached pics of the PSU board. In testing the voltage outputs, I discovered that the +8V output going to the logic board was putting out 14V. To me this is way too high. The PSU logic board rail has two outputs: a +5V and a +8V tap (see IMG_2427.jpg) both going to the processor board.
I am unsure of what the +8V tap was putting out BEFORE the repairs. I also am unsure whether the PSU has ever been touched since its early to mid 90's manufacture date. It appeared that all onboard components were factory default. However I do know that it is now putting out just a tad over 14V (14.1 to be exact) at the +8V output tap.
My question is this, how do I get the 14V close to the spec 8V? Or, is the 14V the right "non load" voltage?
I suspect the answer lies in the LM317K realm somehow. I have found LM317 "calculators" online and have calculated the resistor values that should result in the correct voltage, but they didn't work... either the +8V or 5V were incorrect (frequently both were)
The reading of the LM317K voltages are as follows:
Vin=13.9V
Vadj=3.7V
Vout=4.99V
Readings at BR1 "+" pad is 13.9V
In playing around with different values of R1 (factory value: 75R), R2 (factory value: 715R), and R3 (factory value: 243R) I have been unable to get the +14V down without affecting the Vout voltage. Measurements at R1 are 13.9-14V (the same reading at the +8V output tap on the PSU board). It appears that the 13.99 - 14.1V output is what is coming out of the BR1's "+" pin which is what is being converted from the transformer? Hence, my understanding is that the 14V goes thru R1 and C5 and ends at the +8V output pad (which goes on to the logic/processor board). So another question I have is what does R1 do? Reduce current/voltage across the R1 terminals?
What am I missing?
I would greatly appreciate any guidance, as I am dead in the water at this point, and don't want to put it all together only to fry circuits of the logic/processor board!
Attachments
Last edited:
The LM317K voltages you've provided are correct. The problem is elsewhere.
Is there a PSU schematic available? R1 wouldn't be used to drop the 14V to 8V. You (We) need to trace that 8V section from end to end to try and find the problem.
Is there a PSU schematic available? R1 wouldn't be used to drop the 14V to 8V. You (We) need to trace that 8V section from end to end to try and find the problem.
Thanks sofaspud for your quick reply back!
Unfortunately, I do not have a schematic of the PSU. Theta Digital has been very tightlipped about their schematics to date...
Okay, I have traced the +8V circuit and am pretty confident that I did not miss anything on the board on that particular circuit.
Here it is working backward from the +8V output pad:
1. +8V output pad
2. C5 + pad (470μf/16V)
3. R1 pads (75Ω)
4. C2 + pad (.01μf/100v)
5. C1 + pad (4700μf/16v)
6. LM317k Vin pad/BR1 + pad
For the life of me, I cannot see anything in that +8V circuit that would trim the input/ouput of the 14V at the BR1 + pad. That is, the BR1 is "producing" the 14V that is being outputed at the +8V output pad unabated!
Any ideas? Is the 14V coming out of the BR1 suppose to be the +8V?
I did not include the "trace" on the +5V circuit since that seems to be okay. I have, however, included a pic of the underside of the +8V circuit that I am referring to. (The yellow wire is the +8V output pad.)
Again, thank you for your help and reply...
Attachments
Does the +8 line have a load? If not, I would try 330-ohms 1W or four 100-ohm 1/4W in parallel. It doesn't need to be exact - I'm just looking to draw a few tens of milliamps. The unregulated voltage will drop under load.
Again. many thanks for the laser-fast reply.
The closest I could get (my parts inventory seems to be at an all-time sorry state of affairs--time to contact Mouser!) was 2-680Ω/.6w in parallel. That did bring it down some, but it was less than a volt. I think you are on to something, my friend...
The thought of the load of the logic board circuit bringing the unregulated voltage down to the +8v-ish range did cross my mind. However, I thought that an excess difference of 6V voltage was just too much. Its time to bring out the variac...
I'll go ahead and patch the +8V wire from the PSU to the logic board's +8V pad and slowly bring the AC voltage up while monitoring the VDC going into the logic board. This should give me an working demand voltage going into the logic board at my normal wall AC voltage levels, I would think. Well, at least I will maybe avoid seeing "Old Smokey"!
Many thanks for your replies and help. I greatly appreciate you steering me in the right direction!
I will post my findings so that it may be of assistance to someone else who may have a similar issue with limited knowledge of PSU's (like me). These online forums are just fantastic. It is just like finding many experienced/knowledgeable "heads" from around the world aimed at solving a particular issue. It is people like you that make them work! I just cannot thank you enough!
Cheers!
The closest I could get (my parts inventory seems to be at an all-time sorry state of affairs--time to contact Mouser!) was 2-680Ω/.6w in parallel. That did bring it down some, but it was less than a volt. I think you are on to something, my friend...
The thought of the load of the logic board circuit bringing the unregulated voltage down to the +8v-ish range did cross my mind. However, I thought that an excess difference of 6V voltage was just too much. Its time to bring out the variac...
I'll go ahead and patch the +8V wire from the PSU to the logic board's +8V pad and slowly bring the AC voltage up while monitoring the VDC going into the logic board. This should give me an working demand voltage going into the logic board at my normal wall AC voltage levels, I would think. Well, at least I will maybe avoid seeing "Old Smokey"!
Many thanks for your replies and help. I greatly appreciate you steering me in the right direction!
I will post my findings so that it may be of assistance to someone else who may have a similar issue with limited knowledge of PSU's (like me). These online forums are just fantastic. It is just like finding many experienced/knowledgeable "heads" from around the world aimed at solving a particular issue. It is people like you that make them work! I just cannot thank you enough!
Cheers!
Thank you for the kind words. I only know a little bit about a little bit, but I like to help when I can.
Most of the drop you saw was probably the result of series R1.
Could there be an issue upstream of BR1?
Most of the drop you saw was probably the result of series R1.
Could there be an issue upstream of BR1?
Hi,
Can you read the voltage coming at the yellow wire. It should be 13.9 volt. It should go to R1 and from the R1 to the input of the LM317.
Can you read the voltage coming at the yellow wire. It should be 13.9 volt. It should go to R1 and from the R1 to the input of the LM317.
Thanks for the replies sofaspud and Tauro0221.
Sofaspud, the only component that I can see upsteam of the BR1 is the transformer. On most Theta PSU's, as I understand them, each rail has its own transformers.
Tauro0221, I appreciate your input. Yes, the voltage is in fact in the 13.9V range. But, isn't this too high? Shouldn't it be 8V or thereabout? Does the logic board circuit that will be receiving the PSU voltage consume that much (6V above what is labeled on both the logic board and PSU board? I am confused. I know (thanks to sofaspud for making me realize it) that the +8V pad output voltage is unregulated, but I would think that it being some 6V above what it should be (according to the logic board schematic, I do have that, but no PSU schematic to refer to... Yes, you are correct, the circuit does trace thru R1 and on to the LM317K's Vin pad. So, if I understand you correctly, the 13.9-14V reading that I am getting is normal for the Theta PSU?
Again, thanks for the guidance and replies...
Sofaspud, the only component that I can see upsteam of the BR1 is the transformer. On most Theta PSU's, as I understand them, each rail has its own transformers.
Tauro0221, I appreciate your input. Yes, the voltage is in fact in the 13.9V range. But, isn't this too high? Shouldn't it be 8V or thereabout? Does the logic board circuit that will be receiving the PSU voltage consume that much (6V above what is labeled on both the logic board and PSU board? I am confused. I know (thanks to sofaspud for making me realize it) that the +8V pad output voltage is unregulated, but I would think that it being some 6V above what it should be (according to the logic board schematic, I do have that, but no PSU schematic to refer to... Yes, you are correct, the circuit does trace thru R1 and on to the LM317K's Vin pad. So, if I understand you correctly, the 13.9-14V reading that I am getting is normal for the Theta PSU?
Again, thanks for the guidance and replies...
Hi,
Can you check the resistance of R2 & R3? I think they are the one that set the output voltage for the LM317K. Also the yellow wire is the output from the 14 volt thru R1 unregulated.
Can you check the resistance of R2 & R3? I think they are the one that set the output voltage for the LM317K. Also the yellow wire is the output from the 14 volt thru R1 unregulated.
Tauro0221,
The resistance for for R2 is 713Ω on the dot. Resistance for R3 is 243Ω. You are correct about R2 & R3. R2 bridges Vadj to ground (black wire). R3 bridges Vout (red wire) to Vadj pads of the LM317k.
The interesting thing is that the LM317k' output isn't even in the +8v circuit. The DC comes from the BR1, feeds into LM317's Vin pad/continuing on to the 8V output pad/yellow wire UNREGULATED (the other rails are regulated by LM317T's going to other PC boards). This is why my BR1 + pad reading is exactly what is coming out of the yellow wire AND feeding into the LM317's Vin pad. In essence, there is nothing I can see in the 8V's circuit that will regulate that 14V coming out of BR1.
This is perplexing to say the least. Either the 13.9V reading that I am getting is normal on that particular rail (by the way, this rail going to the logic board is the only one out of spec), OR the transformer is putting out too much DC voltage which would cause the BR1 to output voltage above the 8V spec?. But then, would that also change the LK317k's output voltage?
Many thanks for your input/help. It is greatly appreciated.
Cheers!
The resistance for for R2 is 713Ω on the dot. Resistance for R3 is 243Ω. You are correct about R2 & R3. R2 bridges Vadj to ground (black wire). R3 bridges Vout (red wire) to Vadj pads of the LM317k.
The interesting thing is that the LM317k' output isn't even in the +8v circuit. The DC comes from the BR1, feeds into LM317's Vin pad/continuing on to the 8V output pad/yellow wire UNREGULATED (the other rails are regulated by LM317T's going to other PC boards). This is why my BR1 + pad reading is exactly what is coming out of the yellow wire AND feeding into the LM317's Vin pad. In essence, there is nothing I can see in the 8V's circuit that will regulate that 14V coming out of BR1.
This is perplexing to say the least. Either the 13.9V reading that I am getting is normal on that particular rail (by the way, this rail going to the logic board is the only one out of spec), OR the transformer is putting out too much DC voltage which would cause the BR1 to output voltage above the 8V spec?. But then, would that also change the LK317k's output voltage?
Many thanks for your input/help. It is greatly appreciated.
Cheers!
Hi,
I did some calculation and if you have R3 =240 and R2= 713 then the output of the LM317K = 4.99 it is correct. To output 8 volts you need a resistance of R2 = 1296. Why? Maybe it is a mod to DSPro DAC and they didn't change the +8 label on the board. Did you check the output voltages for UR5,UR4,UR3 and UR2? To see if they are OKAY. A dropped of 8 volts to 5 volt I do not think it will not allow the DSP not to worked. Maybe it is something else.
I did some calculation and if you have R3 =240 and R2= 713 then the output of the LM317K = 4.99 it is correct. To output 8 volts you need a resistance of R2 = 1296. Why? Maybe it is a mod to DSPro DAC and they didn't change the +8 label on the board. Did you check the output voltages for UR5,UR4,UR3 and UR2? To see if they are OKAY. A dropped of 8 volts to 5 volt I do not think it will not allow the DSP not to worked. Maybe it is something else.
I'm confused. Is the rail unregulated? (No lm317). If so, you could drop the voltage with a diode string. But if this is the way it's designed I would be wary of modding it. Is there any where you can find out what the voltages are supposed to be?
Hi,
Can you find out where the yellow wire goes? The only way to find out it is with the schematic for the Theta DSPro DAC. Maybe I am wrong but if all voltage regulators are OKAY then the problem it is in another place within the Dac.
Can you find out where the yellow wire goes? The only way to find out it is with the schematic for the Theta DSPro DAC. Maybe I am wrong but if all voltage regulators are OKAY then the problem it is in another place within the Dac.
I'm pretty sure he has the PSU isolated. It is possible that the silkscreen is the error. vintess, can you trace the +8 on the logic board to a few of the chips? Knowing what they are can help us understand the power scheme.
Problem solved!
In response to tauro0221, yes, you are correct the resistors in the LM317k's path are correct for the 5V ouput (red wire). But remember, the LM317 is not in the 8V's output path. The UR's you mentioned are also not in the 8v's path, as they are for another rail going to another section of the logic board. But, I do appreciate your keen troubleshooting mind/technique. FYI, the yellow wire goes to a section of the logic board (see pic).
In response to cbdb, yes, with the help of sofaspud, I was able to conclude that the 8V rail was UNREGULATED, since no LM317 or similar was in that circuit beginning at the transformer and ending at the 8V output pad. Should I NOT have made the findings outlined below, your suggestion of using a drop-down diode string is brilliant. And, your caution of doing this "mod" is also an excellent observation. As noted earlier in my post, Theta Digital is not a DIY-friendly outfit. There is virtually nothing out there as far as schematics/information on what's inside their components... I greatly appreciate your input/reply...
___________________________________
As I noted at the beginning of this post, the problem of the "over-voltage" has been solved. This is to say, in effect, there was really never any problem. As I stated very early, I did not have a baseline measurement from the in-circuit, powered, 8V pad coming out of the PSU BEFORE my LM317T (not the LM317K) repair and subsequent total PSU recap. And, of course, I didn't have a PSU schematic to initially guide me, either. So, I had no way of knowing whether the post repair reading on the PSU 8V pad was abnormal. Outlined below is what I did.
As noted earlier, my thought was to bring out of storage one of my tools/toys--the Powerstat Variac. It doesn't come out too often, but when it does, it always proves to be a great problem-solver and timesaver!
I have never forgotten one of the lessons my Dad taught me--ALWAYS have the right tool for the right job handy! Anyway, I first tested my AC line voltage at my test bench. SInce we are having a "hot spell" (by SoCal standards) out here in the West and the demand for power has been high, the Fluke tested the ACV out of the wall at 118.4, a tad lower than the normal 121 or so. Knowing this, I hooked up the PSU to the variac. I then hooked up the Fluke to the +8V and ground pads of the logic board (pictured). I also connected a couple of test leads connecting the PSU to the logic board. I knew that my variac's power dial was off by 1 volt. I then started cranking up the ACV on the variac while montoring the DCV at the logic board's 8V pad making sure it didn't get too far past 8 volts. It never came close to 14V at idle I was measuring! At 118.4 ACV, the Fluke was measuring 6.74 DCV's. This was a tad lower than I expected. A couple of things might be the cause of this "low" reading.
It was a relief to me that it was on the low side rather than in the other direction. Intuitively, I did not think that there would be a 6V drop, but then again, it is a large logic board with lots of components that appear to consume lots of juice (see pic). This below spec reading started me looking at the logic board for an explanation. My Peak ESR meter discovered a new issue.
The lone 10,000μf/6.3V electrolytic on the logic board seems to be leaky. So, I now have to order one from Mouser. It is neverending, seems like I am ALWAYS ordering stuff from Mouser! But at least I am now on the right track...
Many, many thanks to all of you who took the time from their busy lives to respond and lend a helping hand. I truly appreciate all of your efforts. Once I get the Theta back together again, I will post further findings here.
Wishing all of you a great, restful weekend!
In response to tauro0221, yes, you are correct the resistors in the LM317k's path are correct for the 5V ouput (red wire). But remember, the LM317 is not in the 8V's output path. The UR's you mentioned are also not in the 8v's path, as they are for another rail going to another section of the logic board. But, I do appreciate your keen troubleshooting mind/technique. FYI, the yellow wire goes to a section of the logic board (see pic).
In response to cbdb, yes, with the help of sofaspud, I was able to conclude that the 8V rail was UNREGULATED, since no LM317 or similar was in that circuit beginning at the transformer and ending at the 8V output pad. Should I NOT have made the findings outlined below, your suggestion of using a drop-down diode string is brilliant. And, your caution of doing this "mod" is also an excellent observation. As noted earlier in my post, Theta Digital is not a DIY-friendly outfit. There is virtually nothing out there as far as schematics/information on what's inside their components... I greatly appreciate your input/reply...
___________________________________
As I noted at the beginning of this post, the problem of the "over-voltage" has been solved. This is to say, in effect, there was really never any problem. As I stated very early, I did not have a baseline measurement from the in-circuit, powered, 8V pad coming out of the PSU BEFORE my LM317T (not the LM317K) repair and subsequent total PSU recap. And, of course, I didn't have a PSU schematic to initially guide me, either. So, I had no way of knowing whether the post repair reading on the PSU 8V pad was abnormal. Outlined below is what I did.
As noted earlier, my thought was to bring out of storage one of my tools/toys--the Powerstat Variac. It doesn't come out too often, but when it does, it always proves to be a great problem-solver and timesaver!
I have never forgotten one of the lessons my Dad taught me--ALWAYS have the right tool for the right job handy! Anyway, I first tested my AC line voltage at my test bench. SInce we are having a "hot spell" (by SoCal standards) out here in the West and the demand for power has been high, the Fluke tested the ACV out of the wall at 118.4, a tad lower than the normal 121 or so. Knowing this, I hooked up the PSU to the variac. I then hooked up the Fluke to the +8V and ground pads of the logic board (pictured). I also connected a couple of test leads connecting the PSU to the logic board. I knew that my variac's power dial was off by 1 volt. I then started cranking up the ACV on the variac while montoring the DCV at the logic board's 8V pad making sure it didn't get too far past 8 volts. It never came close to 14V at idle I was measuring! At 118.4 ACV, the Fluke was measuring 6.74 DCV's. This was a tad lower than I expected. A couple of things might be the cause of this "low" reading.
It was a relief to me that it was on the low side rather than in the other direction. Intuitively, I did not think that there would be a 6V drop, but then again, it is a large logic board with lots of components that appear to consume lots of juice (see pic). This below spec reading started me looking at the logic board for an explanation. My Peak ESR meter discovered a new issue.
The lone 10,000μf/6.3V electrolytic on the logic board seems to be leaky. So, I now have to order one from Mouser. It is neverending, seems like I am ALWAYS ordering stuff from Mouser! But at least I am now on the right track...
Many, many thanks to all of you who took the time from their busy lives to respond and lend a helping hand. I truly appreciate all of your efforts. Once I get the Theta back together again, I will post further findings here.
Wishing all of you a great, restful weekend!
Attachments
I suggest you reform the new capacitor when it arrives.
Do it very slowly using >=100k resistor and leave it to soak for 24hrs at 6.3Vdc
discharge using a 1k resistor and discharge again for about 10 minutes.
Then reform using that 100k for a second time. After 24hours of soak @ 6.3Vdc, you can try to measure the leakage.
The 10M input impedance of the DMM will give a false reading but you can allow for that.
Slowly changing voltage from the regulated 6.3V supply is a bigger problem. charging/discharging through the 100k looks the same as leakage for the DMM reading.
You should be able to get <<0.003CV for your leakage after double reforming.
- Status
- Not open for further replies.