Author Topic: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter  (Read 32968 times)

0 Members and 1 Guest are viewing this topic.

Offline EEVblog

  • Administrator
  • *****
  • Posts: 27193
  • Country: au
    • EEVblog
Dave scored a cheap non-working Stanford Research SR650 8 pole programmable elliptical filter on ebay. Can it be fixed?
http://www.thinksrs.com/products/SR600.htm
Mains Filter datasheet: http://www.jameco.com/Jameco/Products/ProdDS/1174803.pdf
Partial schematic: http://www.ens-lyon.fr/DSM/AGREG-Physique/oral/Notices/N041-023.pdf

! Private video
« Last Edit: May 25, 2014, 05:23:09 pm by EEVblog »
 

Offline dentaku

  • Frequent Contributor
  • **
  • Posts: 830
  • Country: ca
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #1 on: May 25, 2014, 11:51:27 am »
It amazing how much stuff I see here that relates to my history with synthesizers. So much of the terminology is the same.
Of course a programmable filter that's not controllable in real time wouldn't be terribly useful in a musical instrument :)
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 27193
  • Country: au
    • EEVblog
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #2 on: May 25, 2014, 12:09:07 pm »
Could the power supply be suspect and perhaps the relay power rail is not working? Whoever was in there before was in there for some reason.

Yup.
I've noticed a missing heatsink on a 7805 reg, and the 5V rail is playing up. Haven't measured anything else yet.
Looks like there will be a part 2...
 

Offline Shock

  • Super Contributor
  • ***
  • Posts: 2312
  • Country: au
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #3 on: May 25, 2014, 12:55:25 pm »
Dave your video reminded me of some parts I needed thanks!
I instantly spotted the voltage selector (recently swapped a voltage over on one myself) I'm glad it didn't end up too embarrassing :)

If someone has one of those plastic sliding doors please see my post here.

Soldering/Rework: Pace ADS200, Pace MBT350
Multimeters: Fluke 87V, 117, 27/FM
Oscilloscopes: Rigol DS1054Z, Phillips PM3065
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 27193
  • Country: au
    • EEVblog
« Last Edit: May 25, 2014, 05:29:34 pm by EEVblog »
 

Offline VK3DRB

  • Super Contributor
  • ***
  • Posts: 1397
  • Country: au
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #5 on: May 25, 2014, 07:01:42 pm »
Thanks Dave. A good tear down and repair. I get a great sense of satisfaction and achievement fixing something that was junked but was worth a mint. Who hoo! For sure.

I would have thought most mid 90's stuff would have been SMD. But maybe with filters PTH was more appropriate.

Hey, I am in Vichy in France as I write this and wore the negative feedback T shirt yesterday No English speakers around here, but I did notice one chap staring at the message. It appears the Vichy WW2 notoriety has been wiped from memory here. No mention, note even in any museum.
 

Offline Psi

  • Super Contributor
  • ***
  • Posts: 6434
  • Country: nz
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #6 on: May 25, 2014, 07:04:41 pm »
Check the supply rails with the scope.
Might be a supply rail issue causing the random overload states.
The PSU maybe injecting lots of noise/spikes into the analog logic due to its old filter caps.
« Last Edit: May 25, 2014, 07:08:19 pm by Psi »
Greek letter 'Psi' (not Pounds per Square Inch)
 

Online IanJ

  • Frequent Contributor
  • **
  • Posts: 862
  • Country: scotland
  • Pro EE guy many years ago, now a hobby/home biz.
    • IanJohnston.com
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #7 on: May 25, 2014, 08:21:51 pm »
For those interested, there is an SR650 in sorry state from our friends in Israel:
http://www.ebay.com.au/itm/Stanford-Research-Systems-SR650-Dual-Channel-Pass-Filter-SRS-/321057054340?pt=LH_DefaultDomain_0&hash=item4ac07dde84



Ok, I like a wee challenge, have been looking for something to repair/refurb and this will do nicely. Quite a useful bit of kit really.
Am in the middle of a PM3585 logic analyzer repair.......so heck! all this old kit is huge am gonna need a bigger boat bench.

Ian.
Ian Johnston
www.ianjohnston.com
Manufacturer of the PDVS2
 

Offline mikeselectricstuff

  • Super Contributor
  • ***
  • Posts: 11603
  • Country: gb
    • Mike's Electric Stuff
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #8 on: May 25, 2014, 08:29:45 pm »
Looks like it's been damp - could be that PCB track in the filter has corroded away.

Re. the wire into that multiway plug on the secondary - those IDC connectors sometimes go dodgy so could just be it was pulled out of the cable tie to cut down & re-terminate .
Youtube channel:Taking wierd stuff apart. Very apart.
Mike's Electric Stuff: High voltage, vintage electronics etc.
Day Job: Mostly LEDs
 

Offline Tek_TDS220

  • Regular Contributor
  • *
  • Posts: 61
  • Country: 00
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #9 on: May 25, 2014, 08:52:24 pm »
Nice video - I hope that Dave goes through the schematic a bit.  It's difficult to follow because of the way the manual was scanned.

Have these analog devices been replaced with digital signal processors?  Maybe DSP plus the ADC/DAC chain introduces too much distortion for the intended applications?  There's a lot that DSP can do that is very difficult with analog (such as linear phase response).
 

Offline BUkitoo

  • Contributor
  • Posts: 13
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #10 on: May 25, 2014, 10:13:11 pm »
hI!
Regadring the packing, I do not know if it this is the case, but for a product we manufactured once, there were 3M "instant" packing products that were a tiny bag with two compartments filled with two chemicals inside. When you need to pack something you just break a couple of bridges between them and mixed them by hand. Than you place it into the box, your instrument inside, another on the top and the thing grows as a foam inside the bags taking the shape of the instrument.

No machine needed...

« Last Edit: May 25, 2014, 11:22:49 pm by BUkitoo »
 

Offline stranger

  • Contributor
  • Posts: 6
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #11 on: May 25, 2014, 11:28:13 pm »
Nice timing on the IEC input filter discussion. I bought a bit of gear a week ago and the filter started smoking immediately on power up. A lot of smoke and I had to leave the windows open overnight to clear the stink.


Hi Wilfred, seems to be common, I had similar thing with a Schaffner filter input on a Solartron 7150 DVM. It took several months before the failure and even after the smoke cleared there was a big mess inside the instrument as the potting melted around the failed capacitor and leaked out. The cost of a replacement new filter was £25, the instrument cost £20 so I took a saw to the filter and cut the bad stuff out (all of the filter). Dave was very lucky with this filter. More seriously it left me wondering about the safety of some of the old equipment I use and when this is going to happen again - I do have proper current trips on the workbench.
 

Offline ultranalog

  • Contributor
  • Posts: 18
  • Country: pl
  • Audio Nerd - not an audiophool
    • It's only audio
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #12 on: May 25, 2014, 11:53:26 pm »
I thought I noticed a 'chirp' signal source on the HP 35660A. And you said it wouldn't do swept sine? Confused...
playing around with near DC (20 kHz) for fun and profit
 

Offline madires

  • Super Contributor
  • ***
  • Posts: 4281
  • Country: de
  • A qualified hobbyist ;)
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #13 on: May 26, 2014, 12:35:24 am »
Nice timing on the IEC input filter discussion. I bought a bit of gear a week ago and the filter started smoking immediately on power up. A lot of smoke and I had to leave the windows open overnight to clear the stink.


Hi Wilfred, seems to be common, I had similar thing with a Schaffner filter input on a Solartron 7150 DVM. It took several months before the failure and even after the smoke cleared there was a big mess inside the instrument as the potting melted around the failed capacitor and leaked out. The cost of a replacement new filter was £25, the instrument cost £20 so I took a saw to the filter and cut the bad stuff out (all of the filter). Dave was very lucky with this filter. More seriously it left me wondering about the safety of some of the old equipment I use and when this is going to happen again - I do have proper current trips on the workbench.

Most likely that is caused by a bad cap. I heard of a lot of issues with old Schaffner IEC filters with Rifa caps. If the cap fails shorted and if the circuit braker isn't triggered, you'll get smoked potting. That could also explain Dave's burnt PCB trace. A cap failed open releases just some magic smoke. Please see http://www.vintage-radio.net/forum/showthread.php?t=88137 for more line filter issues.
 

Offline madires

  • Super Contributor
  • ***
  • Posts: 4281
  • Country: de
  • A qualified hobbyist ;)
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #14 on: May 26, 2014, 12:45:31 am »
I'm in a similar position. The device was an ANDO AE-5106 communications analyser and it was only $30 so I am not going to source an exact replacement. It isn't even worth the time and petrol returning it to the seller (on Ebay). Interestingly it was advertised as NOS (new old stock) and it looked like it was. So the filter was just old and not worn out. It was probably 20+ years old.The Schaffner name is I think the name on it. IIRC. Also I tried to search on some info and amongst the very little I did find was a post some 6 years ago by someone who also bought one NOS from the same place and had the same problem. I emailed him about some other question, but he had since sold it after removing the filter.

Another one :-( It's caused by a failed Rifa cap in the Schaffner line filter. AFAIK just the age matters, not the hours of operation. If you replace the filter get a brand new one and not an official spare part covered with dust of 15 years.
 

Offline ultranalog

  • Contributor
  • Posts: 18
  • Country: pl
  • Audio Nerd - not an audiophool
    • It's only audio
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #15 on: May 26, 2014, 01:44:49 am »
Can someone explain what Dave was doing when he measured the resistance at the transformer primary and the output of the mains filter with the power on?

Power wasn't on, power button was in the on position, closing the circuit.
playing around with near DC (20 kHz) for fun and profit
 

Offline Khashoggi

  • Contributor
  • Posts: 32
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #16 on: May 26, 2014, 04:09:44 am »
I'm in a similar position. The device was an ANDO AE-5106 communications analyser and it was only $30 so I am not going to source an exact replacement. It isn't even worth the time and petrol returning it to the seller (on Ebay). Interestingly it was advertised as NOS (new old stock) and it looked like it was. So the filter was just old and not worn out. It was probably 20+ years old.The Schaffner name is I think the name on it. IIRC. Also I tried to search on some info and amongst the very little I did find was a post some 6 years ago by someone who also bought one NOS from the same place and had the same problem. I emailed him about some other question, but he had since sold it after removing the filter.

Another one :-( It's caused by a failed Rifa cap in the Schaffner line filter. AFAIK just the age matters, not the hours of operation. If you replace the filter get a brand new one and not an official spare part covered with dust of 15 years.

+1
 

Offline smashIt

  • Contributor
  • Posts: 30
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #17 on: May 26, 2014, 05:49:35 am »
dave, at 15:40 you get a connection to one side of the fuse
but when you briefly touch the other end of the fuse your multimeter didn't beep.
was there no fuse in the holder or is the holder broken?
 

Offline corrado33

  • Frequent Contributor
  • **
  • Posts: 250
  • Country: us
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #18 on: May 26, 2014, 09:01:55 am »
hI!
Regadring the packing, I do not know if it this is the case, but for a product we manufactured once, there were 3M "instant" packing products that were a tiny bag with two compartments filled with two chemicals inside. When you need to pack something you just break a couple of bridges between them and mixed them by hand. Than you place it into the box, your instrument inside, another on the top and the thing grows as a foam inside the bags taking the shape of the instrument.

No machine needed...

Damn, I was thinking of inventing what you just mentioned as Dave was talking about the packaging.

Anyway, good repair Dave. Only you would wish for more of an "electronics" repair.  :-/O Looking forward to part 2!
 

Offline diyaudio

  • Frequent Contributor
  • **
  • Posts: 659
  • Country: za
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #19 on: May 26, 2014, 09:28:44 am »
amazingly one dsp processor can replace majority of guts.
 

Online Rory

  • Frequent Contributor
  • **
  • Posts: 408
  • Country: us
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #20 on: May 26, 2014, 10:34:18 am »
amazingly one dsp processor can replace majority of guts.
DSP not even necessary. LT or Maxim 8th order elliptic switched capacitor filters could do the job just as well.
 

Offline Harvs

  • Super Contributor
  • ***
  • Posts: 1141
  • Country: au
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #21 on: May 26, 2014, 10:47:57 am »
So can these bits of kit do anything you couldn't do with a high end sound card and some filtering software?
 

Offline nitro2k01

  • Frequent Contributor
  • **
  • Posts: 843
  • Country: 00
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #22 on: May 26, 2014, 01:06:19 pm »
Can someone explain what Dave was doing when he measured the resistance at the transformer primary and the output of the mains filter with the power on?
Each side of the transformer is an inductor, so they will pass DC current. He first probed the line and enutral terminals of the IEC socket. Since everything in that whole circuit (fuse, input filter, wires and transformer primary) will happily pass DC if functional, the expected result is some relatively low resistance. The result he got was of course an open circuit, at which he probed the individual parts of the circuit to find where the break was.
Whoa! How the hell did Dave know that Bob is my uncle? Amazing!
 

Offline nitro2k01

  • Frequent Contributor
  • **
  • Posts: 843
  • Country: 00
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #23 on: May 26, 2014, 01:14:01 pm »
So can these bits of kit do anything you couldn't do with a high end sound card and some filtering software?
This unit will pass the DC component of the circuit (if in DC coupled mode) which sound cards typically don't do, by design. It also has zero latency, down to the phase shift of the filter, which may or may not be relevant to your application.
Whoa! How the hell did Dave know that Bob is my uncle? Amazing!
 

Offline jancumps

  • Supporter
  • ****
  • Posts: 1148
  • Country: be
  • New Low
Re: EEVblog #620 - Repair: Stanford Research SR650 Programmable Filter
« Reply #24 on: May 26, 2014, 03:51:34 pm »
So can these bits of kit do anything you couldn't do with a high end sound card and some filtering software?
This unit will pass the DC component of the circuit (if in DC coupled mode) which sound cards typically don't do, by design. It also has zero latency, down to the phase shift of the filter, which may or may not be relevant to your application.
and it goes to 100 kHz.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf