SUN MGA 1200 EXHAUST GAS ANALYSER NEWBIE DUMPSTER CHALLENGE

[obd.tech]

Great stuff guys    thats a great help!.....i will have more of a 'play' tomorrow after work.
That schaffner swiss filter has 91 06 on it....i'm guessing the date of manufacture 1991?
Is it recommended to have one (i may have a donor from the other analyser bench) or best to have a straight input socket/fused switch type?
Will let you know my findings.
Thanks again guys  keep up the good work  that makes this forum

[poot36]

The filter block helps prevent the electrical noise that the unit produces from going back into the power grid.  It may also correct the power factor (PF) of the device.  If you have ever ran a electric motor well watching analog tv or radio you know what the effect of the electrical noise can be.  It won't harm the device to run it with out it though.

[obd.tech]

Post mortem on the blown schaffner 'Swiss Cheese'  picture attached. The dreaded black goo that PAOPBZ posted the forum links about, covers the suspected blown bad capacitor.

I found a replacement in the other gas bench i also saved from the skip/crusher (my planned 'spares' department, newer model SUN DGA1800, but calibration locked out).
This one is made in Thailand, and to my eyes looks externally better made?
Is this one with the possible date stamp of 00 45 (week 45 YR 2000?) likely to have 'short' life caps inside at that age? I will try it as that machine was running ok a week ago.
Any reports of the thai versions going bang? or is it just the 'Swiss Cheese' to avoid? (prefer a nice cheddar anyday    )
More playing continues  ........

[SeanB]

Non Swiss ones are fine.

[obd.tech]

Looked at the transistor/voltage regs? component markings........if you guys can give me some guidance here as i have no idea what this all means (yet.....i'd like to learn)
first 2 are the same             L78S05CV   ST99045           look @ 821jpg
next 1 is(double sqwiggle)   ~/9018       LM317T  P+      look @ 822jpg
suspect?? 1 is                      L7912CV     ST99052           look @ 823jpg
last 1 is                                L7812ACV   ST99050           look @ 824jpg

I'll try and see if i can find some info.......but i 'have much to learn'    but its all good for experience!
Do i take any suspect ones out of the pcb to test again
All help/info welcomed   

[SeanB]

First 2 are 5V 2A regulators, look for a 7805. Next is a LM317, look for that. Third is a -12V regulator 7912. Last is a 12V regulator, 7812.

7805 and 7812 have the same pinout, while the LM317 is a adjustable regulator, and the negative regulator is a different pinout.

[PA0PBZ]

Hmm.... Before you start the fight with the regulators, are you sure there is still something broken?
I would just put it back together and see if it powers up, you obviously found the source of the lightning.
If it does not work start checking the output of the regulators.

[obd.tech]

Thanks....Will have a look?
Any of the readings in the photo's ok or do i need to take out the pcb and put in the $20 'super gadget'?
Not sure about powering it up in the analyser to test the regs voltage on the legs....as don't they need the case for a heatsink?......and i don't want to damage anything else if i can help it guys.

PAOPBZ....thanks......i just wanted to try and learn while it was apart.....and that 7912 in the tester saying faulty...got me worried, i don't want to break the analyser, if it's avoidable, just by being lazy(see that on many cars.....'preventative maintenance is key'...and easier!!.
Remember i have no 'experience' in this, to guide me......give me a car and we are talking different.......thats why i'm here, for help from other people with a passion/interest and enthusiasm/experience in this stuff........and i'm really enjoying it so far 
I'm tempted to go with the 'put back together' option? Let me know.
Cheers guys

[Macbeth]

Is that exhaust analyzer up to current MOT standards? I would love something like that but it does need gas calibration too. I remember the old DIY gas testers you could get fairly cheap in the carburettor and mechanical distributor days, but the <0.3% CO and <1200ppm HC stuff needs pretty specialist and expensive equipment.

Mind you, most of my MOT gas analysis certs had SUN as the device, so I guess it's a goodun!

[PA0PBZ]

You simply can't test the regulators with that device, you have to feed them the proper voltage and see what the output is. The easy way to do that is to put the stuff back together and if it doesn't work... -> measure.

[obd.tech]

Yes Macbeth, this is a Pukka bit of kit, this was from the start of the MOT emissions test days, before it had the computer control 'junk' that now times out and requires calibration @£200?? every 6 months.
This is just as accurate when calibrated as it has the same gas bench as the later version DGA1800 that i also rescued for spares. Tried searching the web.....and world wide government and research departments all seemed to use the MGA1200 for testing. It's really nicely 'industrial quality' kit inside and made in the U.K. (USA parts inside....SUN was part of Snap on company)

She's going back together......

How much heat sink compound on the back of the regs each side of the insulating film? minimal or lots?

[obd.tech]

  I messed up on one of the delicate reg insulator films...man they are delicate...it split 

Switch on time  i got a little nervous 

YIPPEE!!! It LIVES  Yea....well chuffed!!!!!!!!!

Cheers Guys for all your Help   

I've probably been a bit of a div here  .......sorry....a learning curve for me!

First Major hurdle over.......now i can see/find out what works and calibration etc

Thanks again for all your help and input 

p.s. any Auto questions you have....fire them my way (g they are packed with electronics these days)

[HighVoltage]

I have the previous model, the SGA9000 USA Version with 4 Gas analysis.
When I got it, it was broken and dead.
Turned out to be a bad main transformer.

After I got it working, I also replaced all hoses with new pure silicon hoses.
And I cleaned the optics in the IR detectors.

The SUN 4 Gas analyzers have been the best in the industry.
Something like this is not being built anymore.
Just the Infra Red detectors alone are a work of art.

[obd.tech]

Hi HighVoltage 
Great to hear that it's not just me that had a liking to this 'Big Boy's Toy'.
I couldn't bare to see it go for scrap, engineering like this is worth saving.
The pump unit has proper little conrods and pistons 
I guess there are older pukka electronics test kit that the guys on here would appreciate to own for the very same reason. 'Quality Remains long after the price is forgot'  I understand you guy's 
I've seen the 9000 unit...looks very similar....does it have the Andross 257E gas bench?
I'd really like to know how to clean the optics if you could share the info and anything else too 
Mine has HCR (hardware calibration required? on the Co and Hc channels)
Finding info is really rare as this is pre internet kit. But i'm hopeful 
I rescued a couple of spare gas benches too 
Do you have the operators manual in english, would love to get the service manual and schematics too....i'm working on that.
Great forum to share a bit of engineering interest & passion 
Cheers Guys

[HighVoltage]

Hello obd.tech,
I have the full calibration procedure for the SGA9000. It is probably very similar to yours.
It took me "forever" to get this information, but it is only on paper.
Without this paper, you don't even know of how to get in to the calibration mode.

You need a test gas in a bottle and then you can calibrate CO, CO2 and HC.
Butane is good enough for HC calibration.
O2 calibration is done with the oxygen from the air.

I am not sure, who made the gas bench.
But let me take some pictures in the next few days and upload them.
It is a real quality lab equipment and was far too good for typical car repair shops.

[obd.tech]

Hi HighVoltage,
That would be great 
I Cleaned out the plumbing & pump chambers, sorted out better routing and blew lightly across the gas bench ports......mmm.....a fair bit of fine dust came out    not surprising because of the age and also there is no filtration when the unit 'purges' the gas bench for zero calibration.....daft? or maybe job creation for the service guys?
I did have the CO2 channel reading before, but now they are all HCR 
looks like i made it worse 
Gas bench strip & clean required, with the help from you please 

Decided to 'have a play' on the gas bench that was 'in the skip'.
Strip down photo's attached......well nearly stripped. The opto emitter end is easy enough, but i'm stuck on the other end with the motor 'chopper' and receiver sensors?
The 2 motor to heatsink screws didn't do much. Those wires look a little delicate!
what parts need cleaning (emitter end window film +?) and what did you use?
Thanks

[HighVoltage]

Interesting pictures of your MGA 1200
It seems very different from my older unit.
I do have an extra gas bench and will take some pictures tomorrow.

But essentially you just open the optics and clean it all with some alcohol.

[obd.tech]

What type of alcohol?
Is it best if i drink some too 
I just carefully cleaned the ruby mica window by the optical emitter? with a soft cotton cloth and some 'huff'. That mica is pretty delicate!! what are the other optic lenses like?
I look forward to the photo's. Thanks

[obd.tech]

I now have the service technical manual, just need to know the access password to have a  play. Seems the uk versions have a different password 

[HighVoltage]

obd.tech,

I finally got the time to open up an extra bench that I had laying around.
This one came out of a dead SGA-9000 and it looks like it was never opened and cleaned.
For cleaning the tubes and the optics, I have used denatured alcohol in the past and it worked really well.
As long as the good filtering is used, there should not be any dirt coming in to the bench itself.
But over time they do get black.
I am surprised how clean this one actually is.

These gas analyzers work based on nondispersive infrared technology. The infrared radiation is absorbed by the gas
(CO, CO2, HC ect.) and the IR receiver detects more or less of the absorbed transmitted IR energy. The IR transmitter
in this emission tester is just a wound up wire (See pictures). The IR energy is then sent and reflected from a parabolic
mirror and sent through a rotary shutter through the test tube and then it will reach the IR detector. The test tube is
flooded with the test gas and depending on the gas, different wavelength will absorb from the IR spectrum.

These gas analyzers have one of these chambers per gas and they are made for the spectrum of just these three gases.
That is why they have also different length in the test tubes.

I also found an original brochure of this tester, which must be from the early to mid 80th, I would guess.

The PCB seems to be the heater control for the IR transmitter. Looks like the resistors have seen better days.

[HighVoltage]

More pictures

[HighVoltage]

More pictures

[HighVoltage]

More Pictures
This is the original brochure of this gas analyzer

[obd.tech]

Thanks HighVoltage  , interesting photo's    reminds me of an older EGA i looked in some time ago, that had separate channel cells too, i'm pretty sure they looked just like those aluminium ones.
24kg of proper engineering 
This MGA1200 only has the one sample chamber(plastic) in the Andros gas bench, the optic that i cleaned appears to be the collector then? (i said it was the emitter) i'm not sure how to dismantle the emitter & rotor end for cleaning to optics yet.
what was the password for getting to the service adjusting mode?
if i can get that then i think i can adjust the base settings for the analogue bench pcb and i might be operational? This has an analogue to digital converter board and uses a canbus network with the other control boards. Higher tech than i expected.
Was yours a sadly thrown away piece of engineering?
 

[HighVoltage]

I have one fully functioning 4 gas analyzer SGA-9000, almost in like new condition with very little use.
When I got it, it was dead.
Then I bought another one and used it as a parts donor.
It was the opposite, totally used up and bad and also dead, so I just removed the parts I wanted and the case went in to the dumpster. Out of the two, I do got the first one going and it works like new, including the chemical O2 Sensor.

The SGA-9000 only has serial communication, no CAN bus in this old one.

If your only has the one chamber, then it is detecting the different absorption rates of the different wavelength in one single chamber. This was probably not possible at the time, they built the older model.

Hmmm your picture looks like it is the transmitter, based on the heat sink?
I am not sure.

The transmitter should be located across from it of the other side of the tube.
Usually only the transmitter broke. They have a lifespan like a light bulb and moving the analyzer around, while it was turned on, reduced the life span of the transmitter.

I have no idea about the password.
The SGA-9000 has 3 potentiometers for each gas to set the zero, the max and the span.
A complete calibration process of the SGA-9000 takes about 1 hour with different test gases.

I just love the quality of these machined aluminum parts for the bench.
No wonder, the price was over $ 10.000 for the base 2 gas unit in those days.