Sciencemadness Discussion Board

Scientech SA410 Analytical Balance

sbreheny - 1-3-2014 at 17:09

Hi all,

Is anyone here familiar with the Scientech SA410 analytical balance? I have one which I got very cheaply several years ago because it did not work. I'm an electrical engineer and hobbyist so I was able to fix the basic problem easily (it was a power supply problem). I've been using it ever since but it had a few quirks like the displayed value would sometimes suddenly jump by a few milligrams after it had already settled. I figured that this was due to friction and I finally tried to fix the problem.

It looked to me like part of the folded balance beam structure was slightly rubbing against the steel case that contained it. I bent the wall of the case outward slightly in the vicinity of the part that was rubbing and now there are no more sudden jumps so it looks like I got that right.

I also calibrated it using some surplus and new class 1 or better weights.

However, the standard deviation of repeated weighing of the same mass is still larger than spec (I get about 500 micrograms but the spec is 150 micrograms). In addition, if I leave the weight on the pan for a few hours, the reading very slowly drifts around by about +/- 3 milligrams for a 50g weight. This does not seem right to me. This is after sitting ON overnight so it wouldn't be due to not being warmed up for long enough.

Does anyone have any advice on this? Even better, does anyone have the service manual for this balance? I do have the user manual which I used for the cal procedure but it gives no internal adjustment info. There are a number of internal adjustment screws which I do not understand and I did not touch.

Thanks!

Sean


sa410 analytical balance smaller.jpg - 267kB

thesmug - 1-3-2014 at 19:10

I highly doubt this is an easily fixed problem. If you can, find a certified repair technician and ask them if you could borrow their repair manual (or just have them do it :P). I've done this before and they usually don't mind (that is if you give it back!).

[Edited on 3/2/14 by thesmug]

Marvin - 2-3-2014 at 05:53

Quote: Originally posted by sbreheny  
In addition, if I leave the weight on the pan for a few hours, the reading very slowly drifts around by about +/- 3 milligrams for a 50g weight.


This kind of drift has been normal for every electronic balance I've ever used.

For your standard deviation problem I would first check your power supply. What did you fix? Is there any ripple on it? Is the mains frequency the same as the one the filters are set for?

I remember we had a 6 or 7 digit multimeter for a university project that was bouncing around. It turned out it had been in the cupboard so long that the battery had run down. It had forgotten it's settings and defaulted to a 60Hz mode.

sbreheny - 4-3-2014 at 12:03

Wow, Scientech is an AMAZING company - I asked for the service manual for this older model balance which they no longer sell and they sent me a PDF copy, free of charge, knowing full well that I purchased this old balance on eBay. I highly recommend them as a very hobbyist-friendly company.

ZIGZIGLAR - 4-3-2014 at 13:02

Quote: Originally posted by sbreheny  
Wow, Scientech is an AMAZING company - I asked for the service manual for this older model balance which they no longer sell and they sent me a PDF copy, free of charge, knowing full well that I purchased this old balance on eBay. I highly recommend them as a very hobbyist-friendly company.


Testimony to the individual more so than the company. I used to help people out like this when I worked in a technician advisory role for a large distributor - our funding was to assist with warranty repairs, so there was no requirement to spend time on people with redundant products, but I always did. Good stuff.

sbreheny - 4-3-2014 at 13:33

Quote: Originally posted by ZIGZIGLAR  
Quote: Originally posted by sbreheny  
Wow, Scientech is an AMAZING company - I asked for the service manual for this older model balance which they no longer sell and they sent me a PDF copy, free of charge, knowing full well that I purchased this old balance on eBay. I highly recommend them as a very hobbyist-friendly company.


Testimony to the individual more so than the company. I used to help people out like this when I worked in a technician advisory role for a large distributor - our funding was to assist with warranty repairs, so there was no requirement to spend time on people with redundant products, but I always did. Good stuff.


Except that it was a VP who responded to me.

Dr.Bob - 4-3-2014 at 14:33

That kind of variation sounds like most of the older balances I have seen. If you can get it to give you a +- 1 mg weight consistently, that is not bad. I see a drift over time, could be due to temp or even changes in humidity causing the electronics to have slightly different properties. I have even seen a balance that got worse when left on, and a technician said that the older ones can have capacitor and display issues if left on too long.

I have seen the abuse than some balances have taken over the years when labs move, people spill stuff, slam the balance doors, press down on the pan, so there is no telling what might have aged the balance some. But 1mg or better accuracy is good for nearly anything I have done, and can even be built upon by proper serial dilutions*, careful technique and keeping the balance clean.

*ie, if you need a very dilute Soln., first make a stronger one and then take a sample and dilute that-much better than trying to make super dilute solutions in one shot. I have seen people try to weigh 0.7 mg on a 0.1 mg balance to make a soln, which will not give any real accuracy.

sbreheny - 5-3-2014 at 09:43

Quote: Originally posted by Dr.Bob  
That kind of variation sounds like most of the older balances I have seen. If you can get it to give you a +- 1 mg weight consistently, that is not bad. I see a drift over time, could be due to temp or even changes in humidity causing the electronics to have slightly different properties. I have even seen a balance that got worse when left on, and a technician said that the older ones can have capacitor and display issues if left on too long.

I have seen the abuse than some balances have taken over the years when labs move, people spill stuff, slam the balance doors, press down on the pan, so there is no telling what might have aged the balance some. But 1mg or better accuracy is good for nearly anything I have done, and can even be built upon by proper serial dilutions*, careful technique and keeping the balance clean.

*ie, if you need a very dilute Soln., first make a stronger one and then take a sample and dilute that-much better than trying to make super dilute solutions in one shot. I have seen people try to weigh 0.7 mg on a 0.1 mg balance to make a soln, which will not give any real accuracy.


Hi Dr. Bob,

I agree that this balance as-is is accurate enough for home chemistry work. It's just my personality that I enjoy making things as good as I can get them.

I read the service manual and I now have quite a few ideas about what might be wrong. The good news is that I inspected all of the weak points of the delicate weighing structure according to the manual and they all look OK so I don't think it was abused.

It turns out that this balance uses capacitive position sensing of the voicecoil which balances the weigh pan/load. The manual says that the most common cause of drift and lack of repeatability is dirt in between the plates of this sense capacitor and it can be cleaned with IPA and a compressed-gas duster. I will try that in the next few days when I get a chance.

chemrox - 5-3-2014 at 17:26

Scientech, wow. They're a lot better than Mettler or Glas-col. I called Mettler a few years ago trying to get a manual, they were rude. I called Glas-col a few weeks ago and they were just as rude. If you're not a big lab manager with a big budget, fuck you, seems to be the attitude. The analytical "balances" Mettler and Acculab-Sartorius sell today are scales. They run on pressure transducers. I have a pretty good table top one that cost a few hundred. The same transducer in an enclosed cabinet costs around $2k. I'm going to use to try and rehab my Mettler.