September 6, 2017
CRASH!!!! Your 8-channel liquid handling robot arm just raked across the deck and one of the z-axis rods looks bent. No problem, just call the manufacturer and have them come fix it, after all, it is still under warranty...right? Well, maybe...
Most instrument warranties cover parts and labor but, that usually comes with the expectation that the failure is due to normal wear and tear, not abuse or unintended usage. Using the liquid handler failure above as an example, the 8-channel arm likely got damaged because it failed to move to a safe Z-travel height before moving in X or Y. But, was that because the arm failed to execute that command or because the programmer failed to instruct the arm to do so? While a failure such as this might not occur in assays that have been running successfully for some period of time, they are more common when the user is still developing the assay or debugging it. This type of failure could also occur because an operator forgot to retract the arm after some assay interruption or error condition.
Many OEM's (Original Equipment Manufacturer) will work with you to get the instrument back online and some may even be tolerant of such failures to the point of covering the associated costs under their warranty..but, you will most likely find there is a limit to their understanding. If an instrument fails under normal usage, OEM's should and will cover repair costs but if an instrument fails again, or frequently due to operator error the OEM could and should charge for parts and labor and travel, even though the unit is under warranty. Although such a stance would be unpopular for end-users, it is really no different than what you might experience in other areas of your life. If you use your SUV to haul a boat that exceeds the vehicles gross towing rate you will probably damage your transmission or rear axle. Should you expect Ford or GM pay for that? The honest answer is, no.
Whether you bought the instrument new and are under the original warranty, or if you have purchased an extended warranty, make sure you understand just what kinds of failures are covered. Ask up front. Even if you purchase refurbished instruments, there is a limit to they nature of the failures that are covered (BTW - you should always insist on a minimum of a 6 month warranty on refurbished equipment). New or used, a warranty is a quality statement by the provider. Buying instruments "AS IS" or with a "Money Back Guarantee" should set off alarm bells that the low price option that looks so attractive today, could prove to be a costly investment in the future. Caveat Emptor...
What options should you consider when the warranty expires? That will be the subject of our next blog...
July 23, 2013
In the last two installments, we talked about simple DC motors and stepper motors. To recap, a DC motor is basically the simplest (think 'Lloyd' from the movie 'Dumb and Dumber') motor you can find. Simply apply voltage with sufficient current and it spins. You can reverse the spin direction by reversing voltage polarity, and you can control the speed by applying varying voltage levels or by sending short pulses of voltage. Stepper motors are a...'step up' (I hate puns) insofar as intelligence goes (think 'Harry' from 'Dumb and Dumber'). By which I mean, you can send them specific (countable) pulses and the motor will rotate in very predictable increments (steps). Steppers are natively 'open loop' but are oftem fitted with rotary encoders to provide position feedback. However...that position feedback is often an 'after the fact' reconcilliation on the number of steps commanded vs the number of pulses counted. It is not a 'monitor on the fly' feedback loop and that is by definition what a servo motor brings to the table.
In the most basic sense, a servo motor is a brushess DC motor fitted with a feedback sensor. The output of the feedback sensor is used to determine the final position of the motor. These devices can be very small (like hobby RC servos) or larger. In general, with increased size comes larger windings which enable more current and greater torque. Depending on the device the motor is attached to, you can expect to see a series of drive reduction gears or pulleys for even greater torque. Unlike hobby servos which have both gear reduction and circuitry embedded within their housings and use potentiometers for feedback, most industrial servo motors require a separate driver or control board that provides motor voltage and also monitors the encoder in real time. The image to the left represents a simple hobby servo which nicely illustrates the feeback loop concept. Both the input signal and the output from the sensor (potentiometer, in this case) are feed into a comparator circuit. Because the gearing is built in and known relative to the gear ratios, the comparator can essentially decide when the motor needs to be stopped in order to achieve the commanded motion (so many pulses should equal so much resistance).
The pulses sent to the motor are generally all the same voltage but number of pulses sent over time is what determines speed. This is called Pulse Wideth Modulation (PWM).
Troubleshooting a servo motor is not easy. If the motor and encoder are separate units, you can apply the rated voltage to motor (disconnect any pulley/belt or linkages first!!!). Encoders are a bit trickier and we will cover them in a future tutorial. You can try some basics like checking wiring connectors or blowing air the encoder to remove dust, dirt or grease. Beyond that...you will need to crack open the case and hook up a scope to see what the pulse train coming from the encoders look like. Most encoders are off the shelf devices and you can Google specs to compare with what you have. This will at least help you zone in on the encoder or the controller/board it is plugged into.
March 29, 2013
Things break in the lab. Here’s how to protect your equipment, and what to do when it stops working.
With NIH funding fewer than 20% of the research grant applications received in 2011 (the most recent data available) and little hope for improvement in the coming year, researchers must squeeze what they can from every dollar. For some cash-strapped labs, that means buying used instruments instead of new, keeping equipment running long past its warranty, and jerry-rigging existing lab gadgets that might otherwise be scrapped.
When such equipment inevitably fails, it puts yet another strain on already tight budgets. Researchers facing service calls costing hundreds of dollars an hour may feel obliged to delay repairs, only to find that a service tech may not be readily available. Even in the lab, time is money.
There is an alternative. Lab workers armed with a bit of mechanical know-how and some basic tools can sometimes tackle repairs and problems themselves. Not every repair can or should be handled in-house, but those that can will get the lab up and running quickly and cheaply. The Scientist spoke with equipment repair technicians and core facility directors about the kinds of repairs that researchers can and cannot do on their own, and some obvious, but oft-ignored, steps they can take to avoid problems in the first place. Here are their suggestions.
1. RTFM (READ THE F**KING MANUAL)
Like cars and computers, laboratory equipment almost always comes with a manual (either printed or as a PDF). And, as with cars and computers, researchers often toss those manuals in the trash or “file” them in a drawer. Here’s a better idea: as they say on the interwebz, RTFM. Manuals outline recommended maintenance and cleaning schedules, provide troubleshooting tips, and demystify error codes. (Lane Smith, President and Senior Engineer at Phoenix Technical Services, an equipment repair company that serves the University of Mississippi, suggests storing manuals together in a safe place, or as PDFs in a common folder on a lab computer, for easy retrieval.)
The maintenance suggestions these manuals lay out may surprise you. Rebecca Wood, co-owner and vice president of Southern Medical Services, a medical and lab equipment repair firm that serves south Texas, notes, for instance, that some vacuum pump manufacturers specify in their manuals that vacuum oil should be changed after every use. Yet many researchers reuse the oil until it gets dirty—a practice that could potentially cost the lab big bucks. “If you had a vacuum pump that quit and you sent it in for repair and it had black gunk in the oil, they would say you’d voided your warranty,” Wood says.
2. Clean up once in a while
An ounce of prevention is worth a pound of cure, they say, and that’s certainly true in the lab. Dust accumulates on computer parts, ice accretes inside freezers, and carbon dust builds up inside centrifuges and stir plates. Taking care of these issues before they become problems can save a lab some money in the long run. “If you use a [centrifuge] rotor, make sure it’s cleaned afterwards,” says Craig Folkman, a field service engineer atBioNiQuest Lab Services in Danville, Calif. “Make sure O-rings aren’t cracked, and change them as necessary. If there’s a spill, clean it up, don’t let it sit there.”
For mechanical devices that use brush-based (as opposed to induction) motors, such as vortexers and microcentrifuges, Smith recommends investing in a small vacuum cleaner to clean out the carbon dust. (Smith notes that replacing a worn set of motor brushes is a relatively simple task that researchers can do themselves. For one of his techs to make a lab call would cost $200 for an hour of labor plus the brushes ($20–$50 for a set), not to mention travel time.)
Invest in a can of compressed air to clean computer fans and cables, or tracks on liquid handlers. And clean the air filters on lab freezers regularly (they are usually easily accessible on the front of the instrument). “These are fairly expensive pieces of equipment, and some scientists will have their entire research life in these things,” Smith observes. Cleaning or replacing the filters will keep the compressor working properly and prevent it from overheating.
Another easy bit of maintenance: Defrost freezers regularly. “Try to do it once a year, because you will either do it on your own schedule or on the instrument’s—at 2 a.m. on a Sunday morning.”
3. Establish a maintenance schedule
“My feeling about lab repairs is really trying to avoid them,” says Tim Hunter, Director of the Advanced Genome Technologies Core at the University of Vermont. Hunter recommends lab managers ensure that each piece of equipment be kept on a routine maintenance schedule (often outlined in the operator’s manual).
For instance, in his facility, one worker’s job includes tracking the background signal in the lab’s real-time PCR machine, to make sure the instrument is operating correctly. Another bleaches fluid lines in the array reader between runs to minimize cross-contamination concerns.
Another commonly overlooked task, Hunter says, is defragmenting the computer hard drives attached to lab equipment. Hunter suggests doing that monthly. “[These are] things that people take for granted and just don’t check, but [that] can really impact things when you least suspect them.”
Wood recommends copying and laminating the schedule for each piece of equipment and affixing it on or near the machine, so that everyone in the lab knows what needs to be done, and when.
4. Build a basic lab-repair toolkit
Charles T. “C.T.” Moses, an independent consultant in Framingham, Massachusetts, has been offering seminars on laboratory equipment repair throughout the Northeast since 2004. The handout for his seminar includes a suggested laboratory tool set for taking on most basic repairs (see also: www.chastmoses.com/tools.html). To wit: Flashlight; multipurpose screwdriver (slotted and Phillips); small vise grips, needle-nose pliers, and side-cutter pliers; a small crescent wrench; small clamps; scissors; measuring devices (scale, tape measure, or ruler); small hammer; pocket knife; multimeter; polarity tester (to test electrical circuits); and a lockable tool box.
Other items you might want to have on hand are lubricants (e.g., oil or WD-40), compressed air, a set of American and metric Allen and socket wrenches, a drill, and electrical tape.
5. Try the obvious
When something goes wrong, the most obvious solution sometimes is the right one. So, if a piece of equipment suddenly stops working, make sure it actually is plugged in and that the outlet is working; even in the lab, plugs come loose, fuses blow, and circuit breakers trip unexpectedly.
Smith recommends doing a quick and commonsense “self-assessment.” For instance, if the −80 °C freezer is suddenly warming up, did anyone recently perform an inventory during which it was left open? Ditto for the cell culture CO2 incubator.
Next, see if restarting the instrument and/or attached computer solves the problem. Or, if it is a mechanical device, see if you can identify something obvious, such as dirt in a track or hinge, which might be causing the malfunction.
If the problem persists, write down any error codes or messages you see, as well as what you were doing when the problem occurred. For instance, if an autoclave stops working, where in the cycle did it halt? Also, see if you can figure out where in the instrument the problem is occurring. “The easiest thing to do when troubleshooting is to cut your problem in half,” says Folkman. For instance, suppose your HPLC isn’t working; see if you can determine whether the blockage appears to be between the buffer reservoirs and the pump, between the pump and the column, or in the fraction collector.
Even if you end up having to call in a repairperson, such information can save time (and thus money). “If you have an error message, instead of saying ‘I’m getting this diagnostic
,’ if you can say ‘I tried this or that,’ that makes it easier for us. . . . We can know exactly what the issue is.”
6. First, do no harm
If you do decide to open up an instrument to attempt a fix, Smith recommends following the physicians’ creed: First, do no harm. Put another way, make sure you can put back together that which you have taken apart.
Use a cellphone camera to take pictures of wires and instrument settings so you know where they go and how they were arranged. Moses suggests using a piece of white paper, marked to indicate the front and back of the instrument, and taping screws on the paper in approximately the positions from which those screws came, “so you know more or less how it goes back together.” (Oftentimes, instruments may use different screws in different positions, so this kind of information can be invaluable.)
To do no harm also means to protect yourself, says Moses. That means powering down and unplugging equipment before opening it, and putting your left hand behind your back before plugging it back in and turning it back on. That latter point, he says, is an “old electrician’s trick” that “helps prevent shocking your heart should your repair leave a loose wire inside the machine.”
Deciding what can and cannot be repaired in the lab must obviously be done on a case-by-case basis. But as a general rule of thumb, repairs that are covered in an instrument manual’s troubleshooting section can probably be attempted in the lab, such as changing the bulbs on a spectrophotometer or replacing the brushes on a microcentrifuge. Instruments like stir plates are easily fixed, says Moses—often a drop of oil at the point where the motor shaft emerges from the motor, called the bushing, is all it takes.
Instruments that are under warranty or service contract probably should not be repaired in-house, as doing so might void the warranty. Very heavy equipment, very expensive equipment, equipment with precise tolerances (such as a microscope), or equipment involving high voltages, lasers, and so on, probably should be left to experts as well. So should equipment with obvious charring or burning, Moses says, as these might require special expertise in measuring and monitoring electrical circuits.
7. Know when to call for help
When in doubt, it never hurts to call the manufacturer or a third-party repair company. Smith says his company will often offer advice gratis over the phone, and so will most instrument vendors. “I have rarely come across a manufacturer that will not answer simple questions on the phone,” he says.
In some cases, you can purchase a part, like a hinge or a circuit board, and have technical support walk you through the installation process. But make sure the repair doesn’t cost as much or more than buying new. Moses recommends checking the manufacture date on equipment to see how old it is (and thus, if it makes sense to repair it). Circuit boards and electrical components often contain a four-digit code in the form YYWW. For instance, 9613 means the component was built in the thirteenth week of 1996; 0744 means the forty-fourth week of 2007. Though such information will not give a precise manufacturing date, an instrument obviously must be at least as new as its newest component.
But before you do anything, see if you cannot at least identify what is wrong before calling in an expert. By figuring that out you can give the technician the most accurate information, thereby saving him time and you money. Just as practically, you can learn what to do differently moving forward.
“By opening [up a piece of equipment] you can be aware of what happened and what not to do in the future. We don’t want this to happen again,” Moses says.
Originally Published on March 1, 2013 at: The Scientist.com
March 4, 2013
We’ve been hearing a lot of our customers ask about various lab instruments being compatible with Windows 7 lately. Seems IT groups everywhere are struggling with the eventual demise of Windows XP. Already unavailable for new PC’s since 2010, Microsoft has announced that all support for WinXP will cease in April of 2014.
While most instrument OEM’s (original equipment manufacturers) are already making the move to Win7, a huge number of legacy instruments in labs are running XP. Manufacturers may not want to provide ‘backward compatibility’ for older equipment for two reasons; First, if it ain’t broke don’t fix it. Sounds lame, but most instrument software is developed with the OS of the day in mind. Trying to get the performance and reliability that users expect by supporting a major OS upgrade could lead to tons of surprises…ones that they won’t be paid to correct. Also, more than a few vendors have used this Microsoft phase out as a reason to obsolete older instruments and encourage users to upgrade to new hardware in order to get Win7 compliance.
If budgets don’t permit the purchase of new equipment desperate users should consider exploring the Windows 7 compatibility tool. One caveat is that you would be well advised to back up your WinXP first, or better still try installing your legacy applications on a new Win7 PC. It’s a lot easier to mess around with a new PC if you know you can go back to the original PC if all else fails…The following is gratuitously ‘borrowed’ from http://www.howtogeek.com
Using Program Compatibility Mode in Windows 7
It can be quite annoying when you try to install a driver or other software on Windows 7 just to find out it isn’t compatible with the new OS. Today we look at using the Program Compatibility Assistant, and troubleshooting compatibility issues so programs install successfully.
Program Compatibility Assistant
Program Compatibility is a mode that allows you to run programs that were written for earlier versions of Windows. The Program Compatibility Assistant detects compatibility issues and allows you to reinstall using the recommended settings. For example we got this error trying to install a music interface device driver for home recording.
After we closed out of the error, the Program Compatibility Assistant came up advising that the program didn’t install correctly. To try to install it again select Reinstall using recommended settings.
The Compatibility Assistant went through and fixed the issue and we were able to install the driver. The problem was the driver was designed for Vista and the the assistant automatically select the correct compatibility mode for us to install it.
Sometimes you might get a screen similar to this example where Virtual PC 2007 isn’t compatible with Windows 7 and you can check for solutions online.
After checking for solutions online, we’re shown that there is an update that might solve the issue.
Which points us to the Microsoft site to download Virtual PC 2007 SP1.
Note: Sometimes a program does install correctly and Program Compatibility Assistant thinks it didn’t. There are also times when you cancel an installation half way through and it pops up. If you’re an Admin and tired of seeing it pop up because you know what you’re doing, check out our article on how to disable program compatibility assistant in Windows 7 and Vista.
Program Compatibility Troubleshooter
There might be times when Program Compatibility Assistant can’t find a solution, or a program installs fine, but doesn’t work the way it should. In that case you’ll need to troubleshoot the issue. Right-click on the program icon from the Start Menu or in many programs the shortcut icon and select Troubleshoot compatibility.
Windows will detect any issues with the program and you can try to run it with the recommended settings, or go through the troubleshooting wizard. For this part of our example we’ll select Try recommended settings.
This option allows us to test run the program to see if the new compatibility settings fix the issue. Click on Start the program to begin testing it out. After testing the program and determining if the settings work or not click on Next.
If the program is running correctly you can save the settings and it will continue to run with those settings. If it didn’t work properly, you can try using different settings or report the problem to Microsoft and check for an online solution.
If you selected No, try again using different settings it will bring up the troubleshooter where you can specify the issues you’re having with the program.
Depending what you check in the screen above, you’ll be presented with other options for what is not working correctly. Where in this example it shows different display problems.
New settings are applied to the program and you can try running it again.
If none of the compatibility settings work for the program, you’re prompted to to send a generated problem report to Microsoft.
Manually Select Compatibility
Of course if you don’t want to deal with the Program Compatibility troubleshooter, you can go in and manually select Compatibility Mode. Right-click the program icon and select Properties.
Then click the Compatibility tab then check the box Run this program in compatibility for and select the version of Windows from the dropdown. Now it will always run the program in Compatibility Mode for the version of Windows you selected.
Hopefully running the program in an earlier version of Windows helps solve the problems you’re experiencing. Each program is different so the troubleshooting steps will vary. Most programs written for Vista should work in Windows 7, but not all of them. If you’re having problems with a program not working correctly on Windows 7 and have gone through the Compatibility Mode troubleshooter, your best bet is do search the developers website for a newer version or in their forums.
February 13, 2013
With sincere apologies to The Bard, this is a quandry that is often faced by many lab managers when their facilities group or a vendor informs them that a preventive maintenance procedure is being scheduled.
How do you know when the time is right to actually do such work (spend money)? Just because the manufacturer recommends that a PM be done every 6 or 12 months, is that the right thing to do? What if the instrument rarely gets used?
All too often, lab managers or those whose budgets will be tapped for PM services are in the position of ‘erring on the side of caution’ or take a break/fix approach. Spending unnecessarily is obviously not desirable, however waiting till something breaks can cost dearly. There has to be a better way.
A number of common lab instruments have PC based controllers (liquid handlers, readers, integrated systems) and many of those instruments include ‘log files’, which are used by operators to troubleshoot assays or techs to repair instruments. Savvy lab managers and OEM’s can use these logs to track actual usage as opposed to just following suggested time intervals. It requires someone to actually look up the log files (if they exist) and be able to interpret the data but unfortunately there are not a lot of alternatives.
The LabSquad (caution: gratuitous self promotion ahead) is looking for off-the-shelf monitoring solutions that can be adapted to lab use. Other industries commonly use data logging equipment to monitor temperature or humidity but machine usage (especially outside of manufacturing environments) is relatively uncommon. Additional obstacles present themselves in that not all lab instruments use a PC controller and there are not a lot of inexpensive data loggers to choose from. Not to be deterred, we are also looking at custom developed solutions that could be added to any lab instrument which would monitor usage and be inexpensive (cost less than US$100). Just to make it interesting, we would like such devices to wirelessly communicate with a host PC or tablet such that someone could simply pass by a lab like the fellow who reads your home water meter does by driving by your house to assess the usage of key instruments.
While The LabSquad makes it’s living by performing PM’s and repairs, we do strongly believe that we can help labs better spend their support budgets by investing available support funding more wisely. Some instruments (the workhorses) might need more frequent attention, while lesser used devices might have their PM’s pushed out further.
As Paloneus says in Hamlet, Act 2 Scene 2; “Though this be madness, yet there is method in it.” Let us know what you think about PM scheduling and how your lab goes about keeping your instruments ‘research ready.’
February 4, 2013
Although the patent for PCR expired back in 2006 and promised to herald in a new wave of low-cost thermal cyclers, the legal debate over Taq polymerase enzymes continues to make some manufactures nervous about the North American market. Still, the number of new thermal cyclers to hit the market over the last several years has increased dramatically. As the prices for these work horse devices drops accordingly, the justification for service contracts starts to wane. When opting for a low-cost unit with no local service support, some users may be okay with depot repair or flat-out replacement. When opting for higher quality units, many labs are going with periodic maintenance and routine performance rectification (OQ/PQ). Printed reports or recalibrations by the service tech can be incorporated into your lab’s SOP’s but if you are self maintaining, don’t forget to have the data signed off by more than one person, especially if you are doing forensic or clinical work.
Now, let me put my spin on centrifuge support (wouldn’t be a blog without the occasional pun, now would it?). Seriously, it doesn’t matter whether you have a floor mount, bench top or robot-loaded centrifuge, these devices get a lot of use and it is not uncommon to see units that ten or more years old. Motors and bearings don’t last forever so routine maintenance is critical. Additionally, you folks that leave your rotors in the centrifuge and never take them out should have big scarlet letters painted on your lab coats so you can be publicly ridiculed by the service community! Seriously, many a lab tech has pulled a muscle or two trying to loosen and remove a rotor that has permanently bonded with the spindle.
Last on the docket for this posting is microplate reader upkeep and maintenance. Truly, a wide-ranging topic (may have to post separately on this one to do it justice). The three main readers types (modes) are absorbance, fluorescence and luminescence and while some are limited to one mode, others can do more than one (multimode). Of course there are also fluorescence polarization (FP), time resolved fluorescence (HTRF), high content imagers and microfludic analyzers, but for today we will stick with the big three. All three types work on the basic principle of light measurement to detect samples within the wells of a plate. Absorbance readers use a light source, filters and a detector to measure what percentage of the source light is transmitted through the sample. Fluorescence readers are more sensitive and measure the amount of light emitted from the sample, while Luminescent readers have no light source and instead detect a chemical or biological reaction from the sample. Depending upon the specific reader, any number of factors can result in bad data but generally most failures are a combination of optical alignments (emitter, detector, filters…etc) or light source age. Just about every plate manufacturer provides N.I.S.T. traceable “test plates” that can be used to calibrate the device and a number of third-party companies also have more generic standards that can also be used. It seems patently obvious to say, but what is the point of conducting an assay if you cannot say with a high degree of certainty that your detection results are accurate? At a minimum, plate readers should be PM’d once per calendar year and that procedure should include a test report against a known standard. If your lab only has one reader and it is critical to your research, an annual service contract that includes analytical data would be a wise choice.
January 23, 2013
Okay, so we have a blog…big deal. I mean everybody does, right? Just what the world needs. Turns out, there really aren’t a lot of places to go on the web that provide info about maintaining lab instruments. The purpose of this blog will be to try to point out useful resources wherever they may exist in an effort to help folks keep their lab instruments research-ready. Now, since our forte is plate-based instruments, we really won’t be looking at the whole gambit of things you can find in a lab. No freezer talk here. Ditto for water baths, microscopes, MS or GC gear. What we will be talking about are liquid handlers, plate readers, washers, incubators, thermal cyclers…you get the picture. Feel free to chime in and please try to keep it cordial. We are not looking for sales pitches or feature salvos, just sound advice and bits & pieces of useful info.