TC Electronic problems...

[Khron]

The "right" parts might depend on your choice of distributor - not all of them carry all the inventory from all the silicon manufacturers. But since these are simple semiconductors (as opposed to opamps and everything "above" those), the options are indeed multiple.

Is the 2SA1012 in normal TO220-packaging, or the isolated version?

Ordering them by price at Mouser for example, and keeping an eye on the hFE (gain) of the PNP's, 2SA2222G would seem to be an option, as could 2SA2207, both in isolated TO220 packaging. 2N6491 is way down on gain (minimum 20, typical "barely" 150).

IPA70R900P7S, IPA70R750P7S seem to be the cheapest 6-700v MOSFETs at Mouser, rated for 6A and above (0.9 and 0.75ohms Rdson respectively), loads of other options to choose from.

I take it the "F10P04" is also a dual 10A Schottky diode, possibly 40v rated?
MBR2045CT and up (the "45" bit being the rated reverse voltage, going up to 100), SBR1040CT to -10100CT, SDT20100CT etc should do just fine.

 

[drpat]

Is the 2SA1012 in normal TO220-packaging, or the isolated version?

Ordering them by price at Mouser for example, and keeping an eye on the hFE (gain) of the PNP's, 2SA2222G would seem to be an option, as could 2SA2207, both in isolated TO220 packaging. 2N6491 is way down on gain (minimum 20, typical "barely" 150).

The 2SA1012Y is not the isolated version. Was really hung up on if the hFE should be the same, or as high as possible for a SMPS. Stuck the 2SA2222SG non-isolated version in my cart.

IPA70R900P7S, IPA70R750P7S seem to be the cheapest 6-700v MOSFETs at Mouser, rated for 6A and above (0.9 and 0.75ohms Rdson respectively), loads of other options to choose from.

To make things a little clearer, I'm working on two of these power supplies, so these are 6 different part numbers for the same 3 spots (Q1, Q4 and D5). Looks as if I may need to choose different MOSFETS for each of the two supplies based on the Rdson specs. Am I on the right track? Or should I just go for the lowest Rdson (0.75ohms) for both units and call it a day?

I take it the "F10P04" is also a dual 10A Schottky diode, possibly 40v rated?
MBR2045CT and up (the "45" bit being the rated reverse voltage, going up to 100), SBR1040CT to -10100CT, SDT20100CT etc should do just fine.

You are correct... also a schottky. Couldn't find a data sheet for that one either. I originally asked about the RB225T100 because it SEEMS like a more robust part in the same isolated TO-220 package. Could I be looking at this wrong?

Oh, and thank you for holding my hand on this. It's been a fun learning experience, to say the least.

 

[Khron]

Looking at the specs, the "pairs" of components are pretty much interchangeable, no meaningful differences between them.

Pretty sure the MOSFET is the switching element on the primary (and yes, lower Rdson is generally better, but you'll wanna keep the gate capacitance in the same ballpark figure).

Any idea what that PNP does, in the circuit? Is the PNP part of a discrete regulator on the secondary side, or..?

 

[drpat]

Pretty sure the MOSFET is the switching element on the primary (and yes, lower Rdson is generally better, but you'll wanna keep the gate capacitance in the same ballpark figure).

You are correct. The MOSFET (Q1) is located on the primary side.

When you refer to "gate capacitance", the only specs i can find for capacitance on the data sheets for the 2SK1118 and the 2SK2545 are the following:
- input capacitance (Ciss)
- reverse transfer capacitance (Crss)
- output capacitance (Coss)

Any idea what that PNP does, in the circuit? Is the PNP part of a discrete regulator on the secondary side, or..?

I can't find a schematic for this power supply, but the PNP is definitely on the secondary side. The Emitter is directly connected to the standby wire (SB-VCC) that's going from the power supply to the main board. It also leads to the tiny sub-board that contains the PWM chip and opto-isolator chip, but without a schematic, I can't tell exactly where it's connected on the sub-board.

 

[Khron]

That'll be the input capacitance, then. Stay under 2nF or so and it should be fine (stock was 1.3-1.5nF or so, iirc?).

Sounds like the PNP's either a discrete regulator pass element, or just a high-side on/off switch. Either way, higher gain's never a bad thing, it only makes the transistor easier to drive.

 

[drpat]

OK, so here's what I have in my Mouser cart:

(Q1) MOSFET - IPA70R900P7S
Package – TO-220 (isolated)
Type - N-Channel
Drain-Source Voltage (Vds) – 700V
Drain Current DC (Id) - 6A
Drain-source on-state resistance(Rdson) – 0.9ohms (max)
Input Capacitance (Ciss) – 211pF

(Q4) BJT - BD912
Package – TO-220 (Non-Isolated)
Polarity – PNP
Collector to Emitter Voltage (Vceo) - 100V
Collector Current (Ic) - 15A
hFE (gain) – 40~250

(D5) Dual Schottky - RB225T100
TO-220 (isolated)
110V
30A
Common Cathode

Anybody see any issues before I pull the trigger?

I will also update my first post with all of the major parts and part numbers for this SMPS once I test these.

 

[ribteen]

Thought I'd contribute as this thread saved my m2000.

I originally tried the Svart recap to no avail....so I ordered a TDK MTW15-51515 (just fits in the housing) got some jst connectors and a cheap crimper from amazon. I reused the aluminum smps housing bypassed the SB brown wire and left the rear power switch on the bench, thus making the only way to turn it on and off plugging/unplugging, which suits my needs just fine...anyway thanks for all the info here!

 

[fazer]

Fixed the power supply in my M3000. Its back to life. The 50uf/25v cap was not the problem it was on the daughter card. It was the Ctrlr PWM UVL UCC28C42. This is actually easy to check out. You need to gain access to the card and then check the resistance on various pins. If its low in the 2 to 8 ohm range from pin to pin measurements on a resistance scale, its defective and needs replaced. I also replaced the 47uf/400v cap. It measured 22 uf. Its back working and another dopamine fix for the old ego. yeah!

 

[58Glammer]

My M2000‘s PSU was dead..
Did a total recap on it and still bad .. then I exchanged the parts (suggested here in the forum) on the little doughterboard and I still had weird or no measurements on the outputs..
In the end I gave up and put it aside wondering.
About one or two month later after working on recapping some of my other gear, the M2000 still layed around unassembled taking space. My last try was to order a Wizard PSU from a weird Irish shop that said it has them on their website (about 80.- Euros).
They never sent it nor answered requests so I contacted Paypal and got my money back.
So now I had no new PSU and a bad one in the unit I decided to reassemble it and put it in a shelf with other bad gear...
Some of you guys might already smell what went wrong here..
I put the PSU into the housing and was about to just screw it together when I thought... „let‘s see what happens when I connect the defective PSU to the mainboard.. I can‘t get a spareone anyways.. so f..k it, even if it‘s going to explode“
I reconnected all the wires from the PSU, crossed my fingers and switched on the PSU at the back .. and the M2000 .. turns on .. what? I couldn’t believe it.. !!

I (noob) did not get, that in this case the PSU has to be connected to the motherboard in order to work and to get the right measurements
Of course it has to be connected.. there is this second tap powerbutton on the front of the unit that suggest it controlls the PSU
In the end I can‘t tell if it has been the caps (never a bad idea to recap a unit after 20+ years ) or the chip on the small extra board. Maybe both

Anyways I‘m pretty happy it works again. Great unit.. one of my most favorite ones

Thanks for all the tips in this forum!

 

[JohnX]

The 100uf 25v cap that had ESR/short problems was the cause of it all.

I know that I am resurrecting to a very old thread, but I am about to attempt the same repair as described by Svart so many years ago - with a quick question.

After reading through this thread, it appears that I should replace the C7 100uf 25v cap with a 220uf 35v capacitor low ESR. That part seems pretty clear.

This will probably sound like a major noob question but I would prefer to do this correctly once after asking, than make a series of mistakes without asking some specific questions.

I have no trouble locating the capacitor when I’m looking at the surface side. When I turn it over, I may be having a bit of trouble determining exactly which 2 holes need to be de-soldered so that I can remove the old component. Things appear to be squeezed in pretty tightly.

I have attached some screenshots. There is a larger photo of the entire back of the PCB. The zoomed in screenshot has two black arrows pointing at where I think I should be desoldering to remove C7 -
Does that look like the correct location?

Second question - this relates to an email from TC Electronics themselves (first photo) - they had already confirmed the same capacitor replacement as had been mentioned in this thread, but also included this:

“Replace the capacitor with one of equivalent type and specifications or better. We recommend using better "hours" and "temperature" than the one to be replaced.”

Does the 220uf 35v capacitor low ESR as recommended by Svart have the specs regarding “hours” & “temperature” that they are referring to? Or is there something more specific I should look for?

 

[Khron]

There is a larger photo of the entire back of the PCB. The zoomed in screenshot has two black arrows pointing at where I think I should be desoldering to remove C7 -
Does that look like the correct location?

That looks about right, indeed.

Does the 220uf 35v capacitor low ESR as recommended by Svart have the specs regarding “hours” & “temperature” that they are referring to? Or is there something more specific I should look for?

Not sure if he has recommended any particular part number. But you'll want to look for 105C rated capacitors, preferably from one of the major Japanese brands (Panasonic, Elna, Rubycon, Nichicon, United Chemi-Con - one or several are available from all major distributors). You'll want to look through the datasheet of whichever you pick, and make sure it says "low ESR" at the top.

Don't bother with other brands. Just make sure the replacement still physically fits (stock one looks like 6.3mm diameter).

 

[JohnX]

@Khron
Thanks so much.

I am updating my wish list to:

105C rated 220uf 35v capacitor low ESR (Panasonic, Elna, Rubycon, Nichicon, United Chemi-Con only)

I will probably drive into Burbank tomorrow to pick this up and hopefully have it replaced by tomorrow evening.

 

[Khron]

I am updating my wish list to:

105C rated 220uf 35v capacitor low ESR (Panasonic, Elna, Rubycon, Nichicon, United Chemi-Con only)

One thing to keep in mind (which a professional would be aware of) is that there are also 105C-rated "general purpose" electrolytics, which aren't as well suited to the high-frequency high-ripple-current environments of switching power supplies. But low-ESR should be quite clearly in the "banner specs" on the datasheet.

 

[JohnX]

One thing to keep in mind (which a professional would be aware of) is that there are also 105C-rated "general purpose" electrolytics, which aren't as well suited to the high-frequency high-ripple-current environments of switching power supplies. But low-ESR should be quite clearly in the "banner specs" on the datasheet.

In this case, I may be very fortunate in that I will be going to purchase this component in person at a local distributor.
They know me well and they also know that I never really know what I’m doing. They have always been incredibly helpful & knowledgeable.

Any information that I have received here will be relayed verbatim to them. Even if I do not completely understand what I am reading, they always do.

TAW Electronics, Inc. in Burbank, CA.

 

[JohnX]

@Khron I hope I’m not bothering you, but I have two more questions that you may be able to help with:

question 1 - I removed the capacitor at C7, which was not only identified by Svart but also by a YouTube video & a rep at TC Electronics.
But… The capacitor that I removed does not have the value that I was expecting to see and I thought I should just point it out - there’s an attached photo, but it is a Rubycon 47uf 35v 105C

I was expecting to see a 100uf 25v cap there.

This device was sent back to the factory for exactly the same repair in 2009. Is it possible they may have swapped it out with the different value cap that I just removed? There are 3 information sources telling me that I’m in the right location and removing the correct cap so I am a bit surprised that the value was different.

2nd question:
There was a white adhesive (almost like bathtub caulk) which attached this cap to the large one (C5) that has the letter G on it. I had to twist the cap a couple of times to get it to separate from that adhesive after removing the solder.
Do I need to re-add new adhesive there with my replacement?
If so, what would I use?

 

[pucho812]

The adhesive is purely for additional strength to hold the cap in. As gear lives in a rack that usually doesn’t move, you will be ok.
As for cap values. Hard to say without seeing the schematic, well it would be for me. So sorry to interrupt. Carry on.

 

[Khron]

Is it possible they may have swapped it out with the different value cap that I just removed?

Not necessarily, but it's more likely this particular batch of power supplies had a 47u in the BOM (bill of materials) for that part. The value isn't uber-mega-critical (or shouldn't be, anyway), one can expect to see 10-100uF in that position. That, and 47u may well have been a bit cheaper than a larger value. That, and electrolytics have a +/-20% standard tolerance to begin with...

Still, it's gotta be pretty hot and stressful a position, if even a Rubycon YXF can get cooked in there...

 

[MidnightArrakis]

They know me well and they also know that I never really know what I’m doing.

This comment cracked me up!!! HA!!! Well.....at least you're being honest!!!

They have always been incredibly helpful & knowledgeable.

That's an excellent company to do business and develop a long-term relationship with!!!

Any information that I have received here will be relayed verbatim to them. Even if I do not completely understand what I am reading, they always do.

It appears as though you may wish to either take some electronics courses or at least start reading some books and learn about the stuff you don't know and/or understand. Then.....after a awhile, stuff will begin to "make sense" to you.

/

 

[JohnX]

It appears as though you may wish to either take some electronics courses or at least start reading some books and learn about the stuff you don't know and/or understand.

I am fairly sure that all of the information that I’m getting from this forum definitely counts as an education. Everyone in this group has been so helpful and I am far more confident than I was when I first joined. I was able to build my first device just from looking at a schematic recently. I was not able to do that a year ago.

The approach has been to learn something new as each new challenge takes me down a different path. I have been joking that I am learning about DIY electronics one component at a time.

 

[MidnightArrakis]

I am fairly sure that all of the information that I’m getting from this forum definitely counts as an education.

However.....you don't receive any "credit points" as you would get attending a school. Trust me.....I KNOW!!!.....

I was able to build my first device just from looking at a schematic recently.

That's good!!!

I was not able to do that a year ago.

That's called "progress"!!!.....

The approach has been to learn something new as each new challenge takes me down a different path.

And.....the problems begin as soon as when you come up to a "fork" in the path!!!

I have been joking that I am learning about DIY electronics one component at a time.

Every journey begins with that "first step"!!!

[at least start reading some books] -- Here's my "reading material" and how I "taught myself" how to design stuff:



And.....I have gone through and have read (or, at least "browsed") each and every item that you see here!!!

NOTE: The combined weight of all of my "reading material" is around -- 2-TONS!!! (~1,800kg) -- and whenever I have had to move during my life.....you should hear how the moving crews SCREAM TO HIGH HEAVEN about having to lift so many small, yet extremely heavy, boxes from my residences to the moving trucks and then from the moving trucks to my new place!!! But.....I just say to them....."This is what I'm paying you to do"!!!

/