Yet Another Sansui 9090DB Rebuild Thread

[Penn42]

The 53A diode (in the TR09 placement and on the same side of the board as all the blown components) read no conductivity (in both directions). The 56A diode (in the TR10 placement) read .89V. Obviously, neither of those is the 1.8V from post 14. I'm going to order from the German source on the auction site. @DougBrewster, thanks for the reassurance.

I also took out D05 and D06 to test them as I've read they can go bad. When I put each Diode on test I got no conductivity with both lead orientations. I thought this was a sure sign they were bad so I made two sets of two 1N4148 diodes soldered in series as replacements. However, they also test no conductivity with both lead orientations. If I do a diode test on each 1N4148 of a pair individually I get a reading, but when I test across the pair the reading is the same as it was for the original part, which makes me think perhaps the originals are fine. Are these findings correct? Should I go with my new fabricated parts or the old ones?

 

[DougBrewster]

I made two sets of two 1N4148 diodes soldered in series

You did connect them the correct way around ie not anode to anode or cathode to cathode? Or maybe your meter doesn't have enough volts to make them conduct..

 

[Penn42]

You did connect them the correct way around ie not anode to anode or cathode to cathode? Or maybe your meter doesn't have enough volts to make them conduct..

They're connected cathode to anode, stripe to no stripe.

It sounds like there should be conduction and perhaps my meter isn't good enough?

 

[DougBrewster]

If you have another meter you could check the volts out in diode mode using a volts test. Maybe you meter batteries are low?

 

[Penn42]

I've got a crappy little Harbor Freight meter I just used to do that test. My primary meter (an Amprobe AM-510) was supplying 1.48V when on the Diode test function. I have no clue if that's normal or not.

 

[DougBrewster]

I've got a crappy little Harbor Freight meter I just used to do that test. My primary meter (an Amprobe AM-510) was supplying 1.48V when on the Diode test function. I have no clue if that's normal or not.

I don't know if it's low for an Amprobe. My Fluke produces over 5V for diode testung. The 1.48V you're getting won't turn on a 1.8V diode chain

 

[Penn42]

That makes perfect sense. For grins, I pulled the Harbor Freight meter back out. The originals test 1.73 and 1.74 and the 1N4148 pairs tested at 1.69. I'm going to grab a new battery for the Amprobe today and try again.

 

[Penn42]

I pulled off the old and put in all new transistors on F2624 this morning. I tested all the old components and found TR01, 03, 05 and 15 to be bad.

During the process I noticed another inconsistency on the board. R27 has a wire jumper and R28 has what I'm pretty darn sure is a diode. It's not the standard orange with yellow stripe diode found across most of the diode locations. It actually looks similar to a 1N4148. Here's some pics.

Also didn't have any luck with the new battery for the Amprobe. Might be a bit of a downfall of this model.

 

[Watthour]

Ditto that. I recently ordered a small supply of STV3s from his auction listing, and they arrived in several days (probably Luftpost/U.S.P.S.) and appear to be genuine to an ohmmeter. I have no clue how many he has available, so I jumped now before they were gone (hoarding).

 

[DougBrewster]

A diode in place of R28 would cause a big shift in the operating point of TR15 which should be at -1.1V (+1.1V for TR13) whichever way it's connected. It could even be a zener... Why that might have been done I can't guess.

 

[Penn42]

Looking at the schematic from the service manual I'm trying to trace out how R28 connects to TR15 and TR13 and not finding it. Did you mean TR16 and TR14? TR15 was bad when I pulled it, but it was on the side with R27 which had the wire jumper.

I find it interesting that this alien component in the R28 position is in the right channel (even part numbers) and all the bad parts I've found so far, with the exception of R34 & R42, have been on the left channel (odd numbers).

Unrelated: what is the braided nylon tube used to cover leads called? I've tried everything I can think of and can't find the correct stuff on google.

 

[Hyperion]

Unrelated: what is the braided nylon tube used to cover leads called?

I think this is the kind of stuff you are looking for? (I searched on 'braided sleeving' & then 'braided sleeving for electronics'. )

https://uk.farnell.com/hellermanntyton/gss1/sleeving-helaglass-1mm-25m/dp/1674264

Personally I use 'neoprene' sleeving - which is like synthetic rubber, it doesn't burn or perish and it has excellent insulating properties.

 

[Hipocrates]

I've tried everything I can think of and can't find the correct stuff on google.

You can peel a wire and use its insulation too.

 

[DougBrewster]

Looking at the schematic from the service manual I'm trying to trace out how R28 connects to TR15 and TR13 and not finding it. Did you mean TR16 and TR14? TR15 was bad when I pulled it, but it was on the side with R27 which had the wire jumper.

I find it interesting that this alien component in the R28 position is in the right channel (even part numbers) and all the bad parts I've found so far, with the exception of R34 & R42, have been on the left channel (odd numbers).

Unrelated: what is the braided nylon tube used to cover leads called? I've tried everything I can think of and can't find the correct stuff on google.

Yes I did garble that somewhat (it's Saturday night here) I meant the operating point of TR14 should be at -1.1V and TR12 should be at +1.1V. Whatever is in place of R28 would considerably affect TR14 operating point and other stuff around it. The diode version of the board should have a link at R28

The sleeving for which John has provided a link is what I use.

 

[Penn42]

Yes I did garble that somewhat (it's Saturday night here) I meant the operating point of TR14 should be at -1.1V and TR12 should be at +1.1V. Whatever is in place of R28 would considerably affect TR14 operating point and other stuff around it. The diode version of the board should have a link at R28

The sleeving for which John has provided a link is what I use.

Saturday nights get the best of us! Not sure what a link is and a little confused at that sentence in general. Is there a variation on F2624 that has a different number or placement of diodes?

Thanks for all the help.

 

[DougBrewster]

Saturday nights get the best of us! Not sure what a link is and a little confused at that sentence in general. Is there a variation on F2624 that has a different number or placement of diodes?

Thanks for all the help.

Let me clarify a little; the early F-2624 board used a transistor TR09/TR10 to control bias, which was mounted on TR13/TR14. At some point in I think 1997 TR09/Tr10 were replaced by STV3H diodes, R27/R28 were removed and linked/shorted and VR03/VR04 were changed to, I think, 100 or 200 ohms. There was also a change to R25/R26 IIRC. This later board became known as the diode biassed board. Naturally, Sansui never updated their schematic.

There is a third board, which was produced by tnsilver and tolitoms of this forum which is a complete re-make;

http://audiokarma.org/forums/index....-a-modern-remake-of-a-vintage-classic.549021/

The link I referred to is in Hyperion's post number 32 above which deals with your enquiry about braiding.

 

[Penn42]

Good history lesson. I obviously have the diode biasing version then. @LBPete warned me about the 100ohm vr change upthread and I reordered that part.

The new 330mOhm cement filled emitter resistors should be here tomorrow. Hopefully I'll have time to install them and put the unit on test tomorrow night.

Some observations from working on the unit so far:

- I'd read that the pads in these units were fragile and everyone who warned of that is spot on. I've lifted a couple and torn one whilst being very careful. The one I tore I tore with a q-tip of all things.

- The pads underneath TR11, 12, 13, & 14 on F2426 all have a slice cut out of them and I found them much harder to get a nice looking solder joint on.

- Everything seems relatively easy to get to. No required de-soldering of pins is very nice, and having the tuner in it's own tray is very convenient.

 

[LBPete]

The emitter resistors are 0.33 ohms not 33,000,000 ohms. Hopefully it's just a typo.

- Pete

 

[Penn42]

The parts didn't show up on time. I very well could have ordered the wrong resistence if there's no differentiation between "m" and "M". I thought lower case was milli and uppercase was Mega. Don't know how that got into my brain and didn't think to check.

Edit: double checked the mouser listing and I've got .33 ohm resistors coming. Whew!

 

[Penn42]

Got home from trade school tonight and the first thing I did was put the emitter resistors in. Put it on a dim bulb test and it came out of protection (w/ f2624 out)!

That's got me feeling good. I'm waiting on the bias diodes from Germany and a set of transistors mouser had backordered in order to finish F2624. Once that is done and troubleshooted (if need be) it'll be time to hook some test speakers up to try it out. And depending on what I hear maybe even start formulating a game plan for the full restoration.