SX-1980 Problems

Heat sink getting blazing hot. voltages creeping.
Sounds ok but I'm betting it wont last a day.
The creeping voltage makes me think "capacitor".


See quote below my findings
in regards to R212/211 being hidden, I found checking across them to quite easy. Just my two cents.

Meter = Sencore DVM56A

R212 = .77 vdc (.137a)
R211 = .59vdc (.105)
R2 = 26vdc (.055a)
R3 -= 37vdc (.102a)
Note, all voltages are "creeping" up while the unit remains powered on.
r212 started at .75

John aka:hoggwild89

-----Quote------

markthefixer
ESCAPED:armed & dangerous


I have participated in this thread hijack too long now.... time to set things right. (quote this post to preserve the thread flow and information below).

Please start your own thread on your sx-1980 problems.

Read and post the voltage across AWR-154 R210 and then also AWR-154 R212, both are 5.6 ohms, then take the voltage readings in volts and divide them by 5.6 for the current in amps.
Then we'll see if there's an inordinate power draw that is raising the temperatures.

IMPORTANT: The resistors R210 and R212 are hidden, and This thread's post and a few following, with pictures, explains how to access the readings easily.


Since the +/- 34v supply current also flows through R210 and R212, measure the voltage drop for the +/- 34v off board, white cement power dropping resistors R2 and R3, for elimination purposes:
R2 (awr-154 pins 31 to pin 29) (+34 supply) 360 ohms (your voltage) / 360 ohms = ?? Amps
R3 (awr-154 pins 32 to pin 30) (-34 supply) 470 ohms (your voltage) / 470 ohms = ?? Amps

I just haven't gotten a running 1980 and my IR thermometer together at the same time to take heat sink readings (yet).

Searching for posts under my username and containing "R210" (or "r210") yields a few examples of expected normal current. Here's one.
__________________

"Mark The Fixer" is the man you need!

looks like there are resistor numbers not agreeing ..maybe a typo but can be confusing .

petehall347 said:

looks like there are resistor numbers not agreeing ..maybe a typo but can be confusing .

Click to expand...

right, r210 5.6ohms green blue gold gold both usually have fabric shield on them.

I am suspicious because they are both shrouded to some degree under the heat sink.

and these readings don't jibe with his math:

r2 is 360 ohms, and with 26volts across it (26/360) = 0.072A = 72mA
and
r3 is 470 ohms and with 37 volts across it (37/470) = 0.0787A = 78.7mA

so I HOPE it's not Senator John Blutarski taking the readings... "when the germans bombed pear harbor"...



---- so seriously ---------------

if +100 is pulling 137mA total (0.77v /5.6 = 0.137)
and -100 is pulling 105mA total (0.59v / 5.6 = 0.105)

+/-80v are listed at 30mA or 0.030A (0.015A per channel)
while the r2 r3 says 72, 78 call it 75mA apiece and are specified at 65mA, that's not a huge difference

137 - 75 = 62 and if spec'd for 30mA on the +80 , that's 32mA more
105 - 75 = 30 so this side is reasonable.

so, what's pulling harder at the +80? It only goes to the power amps, at 15 milliamps apiece.

each amp has a 15 ohm resistor in series with the 80v input, so it shouldn't be hard to spot.

ohms law says volts = amps times ohms. 0.015 x 15 = 0.225 volts and much more than 0.225v across the 15 ohm resistor is your culprit.

WHOOPS!, Yes, LOL... My math was wrong. Sorry. I am going to double check the readings later.

Here are my voltage readings double checked.

R210 = .65
R212 = .77

R2 = 26
R3 = 37

So I agree on the issue being on the R3 line, the amp boards have been recapped, no rocket science there only 3 caps per board. Now I will go try to isolate the problem unless anyone has another opinion.
Oh, I still think its weird that the voltages are slowly creeping and not stabilizing.

Hmmmm, amp boards,
R107 = .232 on one channel, .236 on the other amp
R105, ok at .212 and .217

taking one board out of circuit at a time made no difference but I did notice a one volt diff between pos and neg, -79 +80 (-79.8 +80.8)
Aside from caps (already eliminated, except for the 22K ufds) I find it difficult to believe I have identical issues on both amp boards but hey... I'm all ears to suggestions.

Just as a stray thought, we have seen leaky ceramic disk caps, and those 80v supplies have their e-caps bypassed by ceramics on the amp boards. quick simple test: pull the ceramics for a few minutes...

Look at the regulator's zener voltage as temp and voltage swing, also look at vbe at the series pass elements on the heat sink.
They are all new? and the 35v "zener" is a 17 plus an 18 zener in series?

What orientation is the chassis in on the bench? Horizontal upside down or is the chassis standing on one end?

It'll take me a few hours, but I'll get the 1980 going and after it heats up, point the IR thermometer at it. I really should nail down a baseline, voltage, temperature, drifts and currents wise...

got some spare 1 ohm resistors to insert into the feeds to monitor the current?

(note to gallery, this guy's a seasoned tech, found out by PM - so this could get DEEP)

FYI, there IS 100-something milliamps cruising through those 5.6 ohm resistors. Made the measurement years ago, and although the exact figure escapes me, I do recall that it was 100+.

Mark,
The unit is standing on its side with the knobs to my left when I use the IR.
I'm not fiddling with this today but on Monday (tomorrow) I am going to attack the beast again re-check all my voltages and currents, then break it down one feed at at time,.
Now, if I am looking at the schematic correctly, R3 is the feed for the -34 volt line and is drawing hot, so Q211 is overheating by default and not necessarily due to the -80 volt line.
Feel free to correct me if I am wrong.
Shoot me a pm if you want my direct email.

Happy Sunday

Readings

With a five minute warm up and heat sink temp about 130 deg F, all current and voltage readings were taken directly (no resistor) using a Sencore DVM56a except for R3. R3 I only measured the voltage drop and calculated the current.
Unless I am missing something, nothing seems that far off.
The eq board is drawing 8ma more than spec so I recapped it,no noted difference after. R3 is higher draw than R2 but I dont really see any notes on what it should be other than what Mike noted.

added note. Found someone replaced r212 prior. measured resistance with a precision meter (above) is 6.7ohms. I subbed in a 5ohm (what I had on hand) and still seems to overheat. (not shocked about that) .586mv and .78mv/117ma and 116ma respectively. The given outputs (of the -80 and -34 rails) total 101ma so I'm thinking 117 is in the ballpark.
This is driving me nuts because I just dont think the heat sink should be able to burn you.

Here is an answer about the regulated power supply's heat sink temperature:

zebulon1 said:

170 degrees F. It doesn't take long to get there, a few minutes is all. You can keep your finger on it for about two seconds.

Click to expand...

That one does run hot, and is in a bad airflow location. Haven't been able to come up with a good way to cool it down YET.

Fans eventually will cake with dust, making things worse. Then there's the space considerations, and lack of airflow through the bottom of the chassis.

Eventually I will add my temperature data from a remanufactured SX-1980.

It would be a lot of trouble, but I was thinking about cutting a rectangular hole in the bottom cover, and JB Welding something like this to the stock heatsink. One on each ear.

What about a remote heatsink mounted on the top-side of the chassis? Or, off the back using existing screw holes?

EchoWars said:

It would be a lot of trouble, but I was thinking about cutting a rectangular hole in the bottom cover, and JB Welding something like this to the stock heatsink. One on each ear.

Click to expand...

My concern in shying away from that, was that a protrusion would take hits, and that turns it into a SX-1010 heat sink situation - busting the board.

The things are in the forefront of my mind are heat pipes. Perhaps based on commercial computer cooling set ups. Wide spaced fins, perhaps with a slow fan - a little air movement can carry a lot of heat.

markthefixer said:

My concern in shying away from that, was that a protrusion would take hits, and that turns it into a SX-1010 heat sink situation - busting the board.

Click to expand...

Yep...I certainly considered that.

The things are in the forefront of my mind are heat pipes. Perhaps based on commercial computer cooling set ups. Wide spaced fins, perhaps with a slow fan - a little air movement can carry a lot of heat.

Click to expand...

What I've seen once or twice was one of the 80V pass transistors moved to one of the output heatsinks and wired with twisted wire. This ends up being somewhat lengthy wire (6+ inches, IIRC), and I dislike such things.

I remember someone who stuck a blower fan on the bottom of an SX-1010 (which suffers from the same issue, albeit nowhere near as badly) and exited the warm air out the back. A slick solution would be something like that on the 1980, attached with magnets (or a couple of small bits of Velcro) to be easily removable, and powered from a wall-wart plugged into one of the switched convenience outlets in the back of the unit. This ought to do a decent job even without messing with heatpipes (who's orientation can be critical to proper operation).

(I linked to one of the first blowers I found on Amazon to show the type of blower I had in mind...I have no idea if that particular blower might or might not be appropriate for the task)

Edit: This one looks interesting: http://www.amazon.com/dp/B004K3BVQ0?psc=1

So then it looks like this is as good as its going to get.

markthefixer said:

Here is an answer about the regulated power supply's heat sink temperature:



That one does run hot, and is in a bad airflow location. Haven't been able to come up with a good way to cool it down YET.

Fans eventually will cake with dust, making things worse. Then there's the space considerations, and lack of airflow through the bottom of the chassis.

Eventually I will add my temperature data from a remanufactured SX-1980.

Click to expand...

It would be interesting to see if the fan motor generated any hash that would be detectable by the pre-amp.

A bigger question in my mind is why is a transistor rated for 8amps generating so much heat with only ~120ma running through it.
heat sinks and fans are like using a tin cup to bail out the titanic. Dont get rid of the water, patch the hole.

Well for shits and grins I am going to let it run for as long as it will run and see if the temp breaches 175deg F. If y'all hear a loud bang then you know what happened.

EchoWars said:

Yep...I certainly considered that.What I've seen once or twice was one of the 80V pass transistors moved to one of the output heatsinks and wired with twisted wire. This ends up being somewhat lengthy wire (6+ inches, IIRC), and I dislike such things.

I remember someone who stuck a blower fan on the bottom of an SX-1010 (which suffers from the same issue, albeit nowhere near as badly) and exited the warm air out the back. A slick solution would be something like that on the 1980, attached with magnets (or a couple of small bits of Velcro) to be easily removable, and powered from a wall-wart plugged into one of the switched convenience outlets in the back of the unit. This ought to do a decent job even without messing with heatpipes (who's orientation can be critical to proper operation).

(I linked to one of the first blowers I found on Amazon to show the type of blower I had in mind...I have no idea if that particular blower might or might not be appropriate for the task)

Edit: This one looks interesting: http://www.amazon.com/dp/B004K3BVQ0?psc=1

Click to expand...

We could make a whole new heatsink and use the existing setup. CNC a chunk of aluminum with fins and incorporate the orginial transistor placements. There is a little extra head room before the bottom cover.The heatsink could extend a good distance where room permits.

hoggwild89 said:

A bigger question in my mind is why is a transistor rated for 8amps generating so much heat with only ~120ma running through it.

Click to expand...

Ohms law...120mA is plenty when you have enough voltage on the transistor. Anyway, the type of transistor used has nothing to do with the heat generated (it'll always be the collector/emitter voltage x the collector current), that's a function of the circuit design. The weakness here is a lame heatsink design, and lame PC board placement.

I have four goddamn 1980's here that need work, and I think I'm going to order one of those fans and find a 12V wallwart and play around a bit. One problem is that the fan I linked to above is 25mm thick, and the 1980 feet are only about 15mm high. Might have to get some hockey pucks to raise the 1980 a bit, or replace the stock feet with these: http://www.amazon.com/Penn-Elcom-91...=8-14&keywords=Penn-Elcom+Rubber+Cabinet+Foot

heat sinks and fans are like using a tin cup to bail out the titanic. Dont get rid of the water, patch the hole.

Click to expand...

Fans work quite well (looks at computer).