Klipsch Bash SMPS DIY Rebuild

C

crazifunguy

New member
Why not start one more project :)
I have a very nice Klipsch Subwoofer that has failed due to poor design in the power supply. Origionally I was going to try and replace just the transformer but after fooling around with it I decided to reverse engineer and redo the whole power supply. Origionally the SMPS used what I believe to be an EER/ER35 Transformer with a vertical bobbin. I have found a good source of ETD supplies at mouser.com so I am planning on re-creating this using one of these cores. I understand that this is not going to be an easy task so I am just starting with the basics of creating the schematic and pcb's.

From what I gather BASH uses self-oscillating push-pull supplies in most of thier amps. Basically any BASH SMPS with a PDC board uses this mode of operation. I have 2 seperate amps/power supplies and they both use the same PDC control board. From this I believe that the ratio of the windings in the transformer are main operational factor.

I will do my best to make sure the schematics and pcb's are accurate as well as component values.
More to come!!!
 
C

crazifunguy

New member
I started with the PDC board 660038RA.


***ALL RESISTORS ARE 1%***
R1 - 200K
R2 - 1K
R3 - 392
R4 - 750
R5 - 750
R6 - 392
R7 - 750
R8 - 750
R9 - 200K
R10 - 200K
(4/5) (7/8) (9/10) are in parallel to obtain desired values.


C1 - R1C 472 KCS 605 (4700pF 60V 5% ????? )
C2 - 3300pF 5% 100V
C3 - 3300pF 5% 100V

D1 - 1N3070
D2 - 1N3070
D3 - 1N5245B
D4 - 1N5245B
D5 - 1N4148
D6 - DIAC (UNKNOWN VALUE)

Q1 - MPS A92
 
C

crazifunguy

New member
I am working on the schematic of the SMPS. Tracing circuits and looking up values is not much fun. I have come to the inductors and I cannot find any data.

How can the inductance of a common mode choke be measured with a LRC meter?

If I measure both sides seperatley I get a value of 8.5mH. Is this an acceptable way to measure this?
That value seems rather large for the common mode choke in the EMI filter.
 
MicrosiM

MicrosiM

Administrator
Staff member
crazifunguy said:
I am working on the schematic of the SMPS. Tracing circuits and looking up values is not much fun. I have come to the inductors and I cannot find any data.

How can the inductance of a common mode choke be measured with a LRC meter?

If I measure both sides seperatley I get a value of 8.5mH. Is this an acceptable way to measure this?
That value seems rather large for the common mode choke in the EMI filter.
Click to expand...


very good job indeed, EMI chock's?

If they are EMI chock's, its normal to get 8.5MH a picture will help more
 
C

crazifunguy

New member
I'll post some pictures right after this.
I finished my first version of the input side of the SMPS. Please have a look and provide feedback. Some of the larger traces could be converted to pours but it should not have an effect on performance to my knowledge.


The origional SMPS had selectable voltage 120/230. Since I live in the US I am only making a 120V version. The only difference I could find is that the jumpers place the caps and resistors in series vs parallel. This also allowed for extra space and ease of routing the traces. I was also able to add 2 large heatsinks for each FET.
 
Last edited:
C

crazifunguy

New member
I have done some searching and reading. I still cannot find if I am measureing the common mode filter properly. I might purchase one with a known value and compare measurements. I also did some component searching. Murara has very similar components in the same footprint but at 2A only 1.5mH
 
C

crazifunguy

New member
I have still been working on the PCB moving some things around. I still have alot of schematic work to complete.....More to come soon when I have the time to tinker!!
 
C

crazifunguy

New member
I have finished my first version of the custom PCB please review and inform.
 
T

TamasM

New member
Hi Crazifunguy,

I have a Klipsch RW12d and the power supply went wrong also. I found a lot's of information in your blog. Many thanks for that!
The PS PCB is different but the PDC board in my power supply. I had to draw down the whole PS to understand where is the problem. (If somebody needs, just send me a mail)
The DIAC type on the 660038RA board is DB6. It is very rare by the way in Hungary...

TamasM
 
A

andrewee

New member
Hi all, I have a Klipsch RW12d and the power supply went wrong also. Fuse was and is not blown. One capacitor 470uf/200v was bad so replaced both. Powersupply is still not starting up. Im figuring out which next steps to take. I cant find schematics for the PDC U402 and also cant find a functional description for this PSU. U402/GND1 is measuring 250 volts, seems pretty high for a gnd... :ambivalence:
 
H

Hemz

New member
Hi,

Thanks for the PCB.
i Do have a Polkaudio PSW 125 and has the same power supply problem.
in fact the input side of the SMPS seems to be badly burnt. as i think the design you created is same as that of my board.

can you please give me the input component value or in face the actual size of the input side PCB.

let me know, if you need more information so i can provide you.

thanks,
Hemz.
 
H

Hemz

New member
Hello all,

Thanks for all the knowledge sharing from this post. I was able to get power out put on the secondary side of the transformer, in fact the Power LED is now glowing..but bang...another problem..Upon connecting all the parts like the speaker and supply input signal from my cell phone for audio. The speaker sounds to very little time and then the standby LED starts glowing and the sound lowers and eventually dies.

The power MOSFET at the output get heatted a lot. so i had to switched it off..

any ideas..

Regards,
Hemz.
 
K

Kenguru

New member
crazifunguy said:
Why not start one more project :)
I have a very nice Klipsch Subwoofer that has failed due to poor design in the power supply. Origionally I was going to try and replace just the transformer but after fooling around with it I decided to reverse engineer and redo the whole power supply. Origionally the SMPS used what I believe to be an EER/ER35 Transformer with a vertical bobbin. I have found a good source of ETD supplies at mouser.com so I am planning on re-creating this using one of these cores. I understand that this is not going to be an easy task so I am just starting with the basics of creating the schematic and pcb's.

From what I gather BASH uses self-oscillating push-pull supplies in most of thier amps. Basically any BASH SMPS with a PDC board uses this mode of operation. I have 2 seperate amps/power supplies and they both use the same PDC control board. From this I believe that the ratio of the windings in the transformer are main operational factor.

I will do my best to make sure the schematics and pcb's are accurate as well as component values.
More to come!!!
Click to expand...

Hi crazifunguy,
Thank you for a nice job.
I have Subwoofer SPF-2 BASH. It is here some problems in the SMPS. My PSU is very similar to yours but not same. I will try to reengineering my design to have the schematics. If you made and have the electrical diagrams of your PSU, would you like please to send them to me or to put the pictures to the site? It will be very helpful. Again, thank you and best regards.
 
R

RichardFaith

New member
Good evening!
Your pictorial and bill of materials was very helpful in my reverse-engineering of a BASH power supply used in a Cerwin Vega CLS-12S amplified subwoofer. The unit was completely inoperative and blew the line fuse immediately upon being switched on. One of the 0.1 microfarad ceramic capacitors bypassing the self-oscillating inverter had (probably shorted and) exploded, and was almost certainly the cause of failure (shorted D,S & G) of one of the two MOSFET's in the inverter. I didn't want to just replace the MOSFET that had failed, knowing that if other problems were present, a whole new chain of events just might cause a new failure, necessitating another tear-down and MOSFET replacement. Therefore, I removed the "PDC" board by cutting all its header pins, and hand-traced a schematic of the PDC board. I did not put reference designators on the hand-traced schematic, because the ones silk-screened on the board do not match the ones in your list; apparently, it was a different revision of the board, although the circuit and all components are apparently identical to the one you worked on. I was able to verify all of the resistors in-circuit, knowing which pairs are in parallel and knowing what other components were in one way or another in parallel with them. By using an external 24V battery and appropriate series resistor, I was able to verify the two 15V zeners. The three ceramic caps, I didn't bother with because I don't have an impedance bridge and anyway that type of cap (C0G ceramic) is hardly at all failure prone. I replaced both MOSFETS, duly considering that the stress event that blew up one of them may also have weakened the other. My on-line MOSFET order from DigiKey was for FOUR MOSFETS (IRF740), to cover any possibility that one or more could have been dead on arrival, and to allow me to try for a matched pair. My method was a little crude: using a 12V battery with a 1K ohm resistor in series, I placed negative on the source and positive on the drain and gate connected together, with a digital voltmeter in parallel with the MOSFET under test. This gives an indication of the forward gate voltage at the edge of conduction. Two of the MOSFETS measured 3.33 volts, one 3.34, and one 3.35. I chose the pair measuring 3.33 volts as the best "matched pair", and they certainly did match much better than I ever expected. Using the pins from a wire-wrap IC socket out of a junk drawer, I rebuilt the right-angle header for the PDC board and put the latter back in place. After cleaning up after all the soldering, I reassembled the unit and it ... ... ... ... WORKED! HALLELUJAH!
By the way, if you would like a copy of my hand-drawn schematic of the PDC board including its interconnections with the MOSFET's, I will gladly send it to you. If you're interested, please give me an e-mail address or arrange some other way I can deliver it. Considering the fact that although Cerwin Vega no longer supports the product, the fact that they still refuse to share the PDC schematic (calling it "confidential") made me quite angry; sufficiently so that I resolved to solve the problem WITHOUT THEIR HELP. I would be more than happy to POST the schematic on this forum if there's a way it can be done.
Thank you,
Richard William Faith
 
R

RichardFaith

New member
The total inductance (in the differential mode; as "seen" across the AC line) would actually be four times the inductance of one side alone, IF the two windings were wound in phase with each other for the differential mode; however, they are not. They are wound in-phase for the common mode, and hence the name common-mode choke. If they were unity-coupled, the net differential-mode inductance would be zero; however, this too is not the case. There exists leakage inductance (i.e., sub-unity coupling) between the two windings, and this is actually by design. Curiously enough, a VALUE for the differential-mode inductance is seldom if ever given. I have not actually tried this, but I believe that if you were to bridge the inductance of a common-mode choke with its two windings connected parallel-IN-PHASE (i.e., with the two INPUT leads connected together and the two OUTPUT leads connected together), the measured inductance would be very nearly the same as that of one of its individual windings. On this basis, I would say that the "nominal" inductance of a common-mode choke equals the inductance of either one of its two coils. Hope this helps. By the way, further down the line I have posted what I had expected to show as a reply to your fine work on the illustration and parts list for the PDC board. Unfortunately, it didn't end up in the expected location. Please look for it and reply. Thank you!
 
You must log in or register to reply here.
Top