EEVblog #621- Stanford Research SR650 Repair - Part 2

Part 2 of repairing the Stanford Research SR650 8 pole Programmable elliptical filter.
UPDATE: This photo from another unit explains all the issues here:

EEVblog Main Web Site:
EEVblog Amazon Store:
Donations:
Projects:
Electronics Info Wiki:

[gglovato]

thumbs up for Dave using the old DS1052E when he has those fancy new DSOs around!

[MrCarlsonsLab]

Hi Dave. I enjoy your videos, and have also checked out the "Live Cam" too.
 I also have some SRS test equipment. Not badly designed, but could have used a little more thought.... IE: my SR780 has the power switch on the back. What the heck were they thinking??
With your SRS filter.... A big part of the regulator heat issue is the missing lid while servicing, on top of the already missing heat sink problem. You can see the air gets pulled in (or pushed) through the side of the case, and would travel right over those regulators. Computer fans push or pull air in a somewhat conical fashion . This would place those regulators in the direct path of air." So they did their homework in that respect. It's a screwy design!
I think EMI and RF noise are the reasons they didn't go for a switching power supply. The cost in added filtering, shielding, and "engineer brain power" was probably worth more than the linear supply they built it with. I'm sure the "bean counters" are a little to blame for design deficiencies.

Take Care :^)
 

[Mythricia]

Might not have been the most in-depth repair video, as you hoped for, but it certainly planted the idea in my head to check for missing heatsinks or keeping in mind heat in general. I am actually really surprised by the behaviour of that regulator - I would not have thought of it as a problem.

But now I know!

[darrenjacobson]

Considering that this thing was held shut with different types of screws, I reckon someone has had this lying around as a parts donor, probably since after the mains filter died. Then at some point they needed a heat sink and grabbed one from in there, that'd be my guess as to why it's missing.

[spiros]

Maybe the high temperatures you get are cause by the no airflow from the fan because the case is open. Try with closed case too

[thomaswilson]

Great video, Dave. Not as complicated an issue as I initially thought. Good lesson to take notice of things like missing heat sinks. Cheers. 

[TimFeleppa]

Good to see the old DS1052E still being used!

[rbarkoch]

Don't even care that it was a simple repair. The repair videos are still great and really interesting. Keep them up!

[BradMorrisKAYAN]

Dave, You (and possibly many of your viewers) may view this as a really simple video, but that's because you have so many years of experience with this type of work.  As someone who has really just begun serious electronics troubleshooting (been tinkering for many years), the simplicity of the failure mode may have eluded many of the "young players" including myself.  I mean...a heat sink?  Geez. 

BTW, thanks for posting the video!!

[Tutoelectro]

If you stop the video in 1:55 and then in 2:07 you will see that it has dimmed gradually.

[GrahamLeach]

Thanks very much for repairing this piece of equipment and demonstrating what it is used for.  I recently got something similar and hadn't the foggiest idea of what it does, how to hook it up, what it might be used for and how to wire it.  Now I am very happy that I included it in lot purchase that I made the other day.  Thanks again.

[ChaplainDaveSparks]

Great bit of detective work on that hardware. It shows how randomly trying to diagnose/fix things often makes things worse. I remember as a teenager many attempts to repair the family vacuum tube television. Back then that would generally consist of removing all the tubes, visiting the local Rat Shack, trying inserting them one by one into the tube tester, and replacing the ones found to be defective.

I'm guessing that someone with just enough electronic knowledge to be dangerous tried to fix this, randomly disassembling parts of it, then gave up and put the cover back on, minus a few components. Removing a heat sink, though, is the work of a neophyte. It's just like trying to fix an old single cylinder gasoline (petrol) engine by first adjusting the carburetor, which is usually the LEAST likely thing to be out of adjustment.

[vex]

I'm glad watching useful videos Dave! :)

[OneBiOzZ]

2 things, for one is the falloff better? 
and two, the 7805 had a silpad on it and the TIP did not and they were right next to each other were they mounted on the same heatsinc? make sure the TIP is not overheating! 

[frankschieschke]

Hi Dave,
i suggest that the 12V fan is connected to the +15V and the +5V.
That is the reason for increase the RPMs while the +5V dropped.
I also suggest, that the original heatsink was like a L-shaped sheet of aluminium,
which is screwed to both the 7805 and the transistor and has cutouts for passing the fan blow through. In the upper corner you can fix the sheet to the fan holes.
Frank

[SiorusBzns]

Couple years late to the party here, but with respect to the thermal performance:

Remove the stamped fan grill, replace it with a wire grill, and flip the fan around. Airflow volume isn't _directly_ related to heatsink performance; airflow velocity is. The two values are (obviously) related, but the latter is ultimately what affects heatsink performance.

Running the fan as an intake will generate more directed airflow over that power supply section than running it as an exhaust would (by virtue of the fact that axial fans tend to pull air in from every direction while the output is at least somewhat collimated) and produce a significant decrease in operating temperature on those voltage regulators. I don't think this thing uses enough power for the fact that you'd be pushing pre-warmed air across the mainboards to matter; I don't really expect the temperature rise to be meaningful.

In fact, if you ditch the existing grill and reverse the airflow direction you can probably run a fan that's rated for perhaps as little as half the airflow and static pressure of the current one and still improve thermal performance. I'd recommend something from Sanyo Denki; they're of markedly higher quality than EBM-Papst or Delta and they tend to be quieter for a given flow/pressure curve. Nidec would be excellent, too, but they appear to be hard to come by, at least in the states. Mouser and element14 both carry Sanyo Denki (not sure about australia-specific distributors).

[eurias-nl]

(Note ADVERTISEMENT !!!! )

To make a comparison with a competing product, as a supplier of used equipment, I can wholeheartedly recommend the Kemo VBF8. That model has approximately the same options in terms of filter characteristics. Their gain is only 0 or 20 dB, and the channels have a common power supply. Additionally the filter slope can be reduced to a Bessel-like "pulse respons" shape.

But their power supplies do not have the disadvantages mentioned in this video. The power supply of the VBF8 is designed with a toroidal transformer and such a two-channel instrument consumes no more than about 13 - 15 Watts. And because they contain no LED displays or microprocessor, the VBF8 does not require a fan, and adjustment of cut-off frequencies works with mechanical rotary switches.

For those who want to use such a filter instrument, Eurias can supply a number of units that were assembled in the previous century. With minor signs of use, no rusty connectors, calibrated and tested, they still function as well as when they were first put into use. In addition to variants with an elliptical response, there are also some with an 8-pole Butterworth response. And they are supplied for connection to a 120 VAC or 230 VAC mains voltage, to be specified on order.

[behrensish]

I really enjoy watching your videos. You re-ignited my interest for EE. Too bad I spent so many years on computer Science...

[ishcosplay]

Hey Dave, don't put the repair down. I think this is a great example for a beginning service tech in any field. I've been a tech for over 30 yrs and plain observation is a hugh part in repair of any item. As soon as you noticed wrong screws in the housing, you should always expect anything. Tamper damaged equipment can be a nightmare to repair so rejoice in your ability to find a solution when someone else couldn't.

[redtails]

5:30 wow.. it's amazing that someone noticed this in the previous video, i would never have