To contact us for support of an RKM Engineering supplied product or system, please send an email to our support group.
RKM Engineering has many service contracts in place. Our contract holders enjoy priority support for their system. In addition we perform non-contract service. Contact us for current rates.
Registered clients: please click here to download upgrades, etc.
Frequently Asked Questions
Note: this FAQ about our products will be updated regularly with newly received questions, as well as issues from our archives.
Yes. EIS2002 requires an upgrade to version 4.02 or higher to work correctly with the 2007 DST changes. In addition, the computer on which EIS2002 is installed requires an MS patch (see MS KB 931836).
Under some conditions, this database file can grow large, depending on the type of reports being generated. It is a good idea to regularly compact this database. To do this, first close EIS2002. Then launch EIS2002 as EISADMIN, then go to File | Open and open EIS2002_wkstn.mdb. Then go to Tools | Database Utilities | Compact and Repair Database. When this returns, close this database, and confirm that the size has decreased. An alternative way of opening Eis2002_wkstn.mdb is by using the "Secure Access" shortcut, if such a shortcut has been placed on the desktop.
remove the security prompts on Access2003, go to Tools | Macro | Security, then
select Low Security.
Generate the graph, then click on the "edit chart" button. This is the button with the notepad and pencil icon, in the top right section. Then click on the "general" tab, and then on the "export" button. You then can save the graph in your favourite format (jpg is probably most appropriate). Then in word use Insert Picture, and select the file you just saved, or paste it if you clicked on "Copy to Clipboard" instead of "Save to File".
When I print a colour graph to a colour printer, it prints in black and white. How do I print in colour?
EIS2002 forces all colour to black to make things work better on monochrome printers. To print in colour check the "use a colour printer" check box on the graphical form.
The following assumes that your EIS2002 application is located in the "E:\EIS2002\" folder, and that your backups are stored in the "F:\EIS2002 backups\" folder. These locations may differ between installations.
Database corruption usually becomes clear when the daily compacting process becomes unsuccessful. It is important to realize that there are two distinct databases being compacted and backed up: E:\EIS2002\eis2002_dta.mdb (which holds all short term and hourly average data) and E:\EIS2002\eis2002_be.mdb (the configuration data base and all other data). The last uncompacted versions are saved in the F:\EIS2002 backup\ directory as eis2002_be_prev.mdb and eis2002_dta_prev.mdb, but only if the compacting was successful. The newly compacted versions are also saved in the F:\EIS2002 backup\ directory, as eis2002_be_back.mdb and eis2002_dta_back.mdb (the "backup" versions), again only if the compacting was successful. Please note that I will be reviewing this in the future, as my original intent was to keep the previous day's copy instead of a compacted and non-compacted version of today's database.
As a result, the databases found in the F:\EIS2002 backup\ directory can always be expected to be uncompromised. To restore, you must first close all EIS applications (supervisor, poller, eis2002, repsvr). Next check the dates on both "backup" versions. If they are the same, then it is best to restore both of them (by copying "F:\EIS2002 backup\eis2002_dta_back.mdb" to "E:\EIS2002\eis2002_dta.mdb" and copying "F:\EIS2002 backup\ eis2002_be_back.mdb" to "E:\EIS2002\eis2002_be.mdb").
However, in most cases only one of the two databases failed to compact. As a result, the good versions may not be of the same day. The action to be taken depends on which file failed to compact.
If the _be file failed, which is indicated by the _be_back file being older than the _dta_back file, you should only restore the eis2002_be.mdb file.
If the _dta file failed, which is indicated by the _dta_back file being older than the _be_back file, you should only restore the eis2002_dta.mdb file. But following this you must set back the "last valid rtu time" for each RTU to a time and date well before the time and date of the eis2002_dta.mdb file, in order to minimize data loss. To do this, open the EIS2002 application as EISOP, cancel the report scheduler, and go to the Poller Configuration. Close EIS2002 after that.
Regardless of the way you performed the restoration, it is a good idea to verify that compacting will now work. To do this, open eis2002 as EISADMIN, go to Maintenance | Compact on Demand | Compact Now. If it appears to work (it will take a while) the restoration was successful. Close eis2002, then restart the computer.
The number of exceedances as reported by the EIS Monthly Report does not always match the number I count on that report. Why?
The hourly values shown on the Monthly Report (and all other reports) may have more precision than what is shown. The number of decimals used is determined by the Parameter Type Configuration, while the precision of the value as reported by the AQDL RTU is determined by the Engineering Unit Equation, as set in the Parameter Configuration. AQDL2 RTUs report the data in floating point, with near infinite precision. So the values reported by the RTU and stored in the EIS can carry additional precision over that shown in the reports. Even if the precision at the RTU level and the number of decimals shown in the report match, the data will still have additional precision if the report is based on running averages, e.g. 24-hourly running averages. As a result, when a value of 170 is reported, it may actually be as low as 169.501, or as high as 170.499. If the criterion for exceedence is set at 170, then a reading of 170.1 would show as 170, with the "Decimal for Reports" set at 0. Similarly a reading of 170.6 would be reported as 171, but both values would be included in the count of exceedences, as both are greater than 170. But a value of 169.9 would also show up as 170, but it would not be included in the count of exceedences.
The timestamp always refers to the ending time of an average interval. Thus hourly data stamped 2000-12-01 00:00 is acquired over the hour starting at 23:00 on November 30, and ending at midnight. Thus the 00:00 reading is the last value for the previous month.
As explained above, timestamps always refer to the ending time of an average interval. Thus the first column of each day is for the period ending 01:00. Therefore we label it hour 1. The last hour could be labeled 24 or 0, depending on your taste. We prefer 24.
For each field on a MAP display, a "deadtime" in minutes is defined in the Map Configuration record for that field. If the difference between the current timestamp of the data displayed in a field and the current time is greater than the defined deadtime, then the background colour of the field will turn grey, until the condition is removed. Thus one can always trust that data is current as long as the field's background colour is not grey. For wind vector displays, the vector is simply blanked out when the data is not recent enough. It is easy to see the time stamp of a field: just hover the mouse over the field and the timestamp will be displayed. In some cases, the deadtime may have been set too short, in which case it should be adjusted.
Typically this happens when the Associated Wind Station and Channel have not been set correctly in the Parameter Configuration for the affected parameters. To correct, go to Configuration Functions | Parameter Configuration, and locate the affected parameter(s). Select the nearest Station which has a wind parameter by clicking on the dropdown button for "Associated Wind Station", then select a wind direction channel by clicking the drop down button for "Associated Wind Channel". Finally save the changes.
WindowsNT systems running EIS2002 are usually set up for auto logon. As a result, on reboot, the user will not be prompted to enter a username and password. When it is necessary to log in as Administrator, e.g. in order to install a printer, then the auto logon must be by passed. To do this, press ctrl-alt-del to bring up the Windows NT Security panel, and click on "change password". Change the default user's password, then logout. You will now be presented with a message that NT could not log you on, and you can enter the Administrator's user name and password. After the necessary changes have been made, you must once again log in as the default user, using the temporary password. After that, change the password back to the one used by the automatic login process.
When a user forgets to log in as default user after having logged in as Administrator (see above), the default user value in the registry will be set to Administrator. Correct this by logging in as default user (normally "EISUSER"), and the next auto boot sequence should work. If it does not, then the user may have forgotten to set the password back to the correct auto logon password.
I am using the EIS2002 features which allow me to automatically send alarm messages and/or automatically generated reports by email to various destinations, but I get prompted to acknowledge this. Now what?
If you run Outlook98 upgraded with Outlook Security, or Outlook2000 or Outlook2002, the new security features have "broken" programmatically created email messaging by inserting the prompts, when these programs use CDO (see MS KB article Q262701). EIS2000 has now been rewritten using MAPI instead of CDO, resulting in correct automatic messaging without prompts. This is implemented by EIS2002 v3.02, ByteArrayToSng.ocx v3.1 and Mpol.exe v188.8.131.52. You will need to upgrade to these or later versions to eliminate the prompting.
My Alarm History log that is displayed at the bottom of the EIS page is getting very full. Is there a way to erase or archive some of the past alarms?
You can delete and or archive alarms very easily, in the same way as you do for data. Log on to EIS as EISADMIN and go to Maintenance Functions | Backup, Purge and Restore, then select the appropriate time range, select "Alarm messages" and click on "Backup" to save the messages in a database, and/or click on Purge to delete them. Please note that backing up does not remove any records, you have to expressly purge them after any backup.
The way the alarm system works is that it looks at the most recent reading received, and compares it with the previous last reading stored. E.g., if you want to see alarms for individual 5-minute periods, then you have to speed up the poller rep rate to 5 minutes.
All data is time stamped using standard time. During the summer, the EIS2002 computer will run in Daylight Savings Time, but the data will still be reported in Standard Time. Even if you time synchronize the RTU from EIS2002 during the summer, the RTU will be set to the correct standard time.
It is very simple to use the AQDL2 on-board sensor as a linear recorded channel. The Register Assignment in the EIS2002 Station Configuration must include the AIN 5203/4 Temperature module. If the register assigned is 30017, then a parameter can be configured for this temperature, by setting the physical channel to 30017, scale to 32767 for AQDL2, 16384 for AQDL2-LP, offset and dp to 0, and allocating storage for it. Temperature reading will be in degrees C. Set the physical channel to 30018 to record in degrees F.
This error indicates that the background process which acquires data has found a wind speed channel without matching wind direction, and it occurs when the wind speed channel is downloaded before the wind direction channel information has been sent. The problem only occurs if the Parameter ID of the wind speed is lower than that of the direction. This problem can be ignored unless it is recurring, in which case the matching wind direction problem should be corrected. For a complete list of all AQDL2 error codes, refer to the "AQDL2 Protocol Description" manual or to this AQDL2 reference summary.
This depends on whether or not the EIS server is time synchronized automatically. If the server is part of a network, time synchronization is typically maintained by the network. Stand-alone system that are connected to the internet may be automatically time synchronized as well. At any rate, if the server can be expected to keep proper time, then it is best to automatically time synchronize the RTUs once a day. This
AQI Configuration is used to configure all sorts of computed channels: actual AQI (Air Quality Index) parameters as well as block and running average parameters. The "pollutant" field has meaning for actual AQI parameters only. However, leaving this field blank for non-AQI parameters will prevent new records from being added. Thus this problem can be solved simply by selecting a value for the "pollutant" field. As of EIS2002 v3.37 this field defaults to a non-blank value, preventing this problem.
This problem occurs if AQI break points are entered normally, but the unnecessary fields are left blank instead of set to 0. AQI breakpoints are parsed starting at the first one until the value 0 is encountered. Thus this problem is simply solved by entering 0 on the first unneeded break point field.
There may be a number of reasons for this, but a simple cause may be that the first AQI breakpoint is set to 0. AQI breakpoints are parsed starting at the first one until the value 0 is encountered. Thus this problem is simply solved by entering a non-zero value in the first break point field.
The Custom Filters form is displayed in modal mode, which implies that the toolbar delete button is not enabled. To delete records from the Custom Filters form: move to the record that you want to delete, then click the record selector bar (thatís the bar along the left hand side of the form, then hit the Delete key. Youíll be prompted to confirm that you want to delete the record.
What is the reason that on a new EIS2002/EIS2008 installation on a new workstation, everything appears to work, except that I cannot generate any reports other than Browser reports.
A possible reason is that there is no default printer installed. Go to Start | Printers and Faxes, and install an available printer.
After upgrading from EIS2002 to EIS2008 my stations appear sorted by name instead of by station number. How can I change that back?
In any of the forms, e.g. Station Configuration, click on the Station ID field, then click on the A-Z Ascending Sort button on the Home ribbon in Access2007. The new sort will stick and can be undone by clicking on the Clear All Sorts button.
This problem is due to MS security update KB960715 and more recent cumulative updates. This problem can be corrected by downloading a small program from the user download page.
I corrected the Browser problem (see above) by rolling back KB960715, but as of July 2009, the problem has returned. Now what?
This problem is due to MS security update KB973346 and possible future cumulative updates. Rolling back the updates will not be a lasting solution. Instead the problem should be corrected by downloading a small program from the user download page.
In most applications the temperature sensor is used to flag or invalidate data outside of temperature limits which can be set in the station configuration. It has always been possible to record the temperature as well, as it is available as physical channel 11, but this would result in a non-linear result. As of AQDL version 1.60 (June, 2000), the AQDL RTU can now record on-board temperature in a linear fashion. To use this, set up a channel using physical channel 11 and set the treatment to non-linear. With the scale set to 4096, offset 0, and dp 2, the result will be the board temperature in degrees C. Users with an older version can upgrade to the newer version to enable this feature.
From EIS2002, use the Poller Configuration, Connection Tab, and drop down the box labeled "time synch". Select the time synch mode, typically "time synch on next RTU access", and save the change. When the RTU receives such a time synch, it moves the data up or down accordingly, so make sure that the EIS2002 computer's clock is set correctly.
On an AQDL RTU, you can also set the date or time directly, using the diagnostic mode. This does not affect the existing data.
Yes, if the instrument's serial port is connected to COM1 of the RTU (the RTU port which is normally used for local diagnostics). When you call in through the RTU's modem (COM2) using DVTERM, switch over from MODBUS to ASCII protocol if necessary (as documented in the AQDL User's Manual). When you press <enter>, you see an extra item at the prompt line, named "port". Select this function, and you will be connected through to the instrument on COM1, until you press <ctrl-x>. The baudrate and format must be the same for the connected instrument and the diagnostic port.
Alarm levels are common between short term and hourly readings for AQDL data loggers. This has been improved for AQDL2 data loggers. There are ways to get around this though. You can define an additional logical channel at the RTU level, pointing at the same physical channel, and you can set the alarm levels different for the extra channel. You will have to define a new Parameter Type for that channel as well, as each Parameter Type can be used only once for any station. Or you can use the AQI scheme to compute a new channel, with its own alarm level. The latter method is somewhat smarter, as it doesn't use up RTU memory space. For the latter method an additional Parameter Type is to be defined as well.
With a multi-parameter instrument such as a NOx analyzer with outputs for NO, NO2 and NOx, with autocalibration set up for NO only, there are two possible ways to invalidate the non-calibrating NO2 and NOx channels. One way is to use the zero and span bits to flag the data. That works OK, except that the data during the recover phase will not get invalidated, so you'll get a bitof a higher reading at the tail end of the calibration. The other way is to set up a calibration schedule for NO2 and NOx identical to the one for NO. This works fine as long as you don't use the abort level (which postpones a calibration if the last short term average was too high). One needs to set the abort level high for all three channels. Please note that the AQDL2 auto-calibration scheme allows for using the first method for all calibration phases, including any recover phase, without compromise.