Fitting a lamp with no RFID chip, and what you have to do yourself

Several deuterium lamps are sold in two versions, one carrying an RFID tag and one without. The tagged version costs more, and the price gap on a single lamp is real money.

So the question comes up on nearly every quote. Will the cheaper one work?

On an Agilent UV detector the answer is yes, but there is a setting you have to change first. At first it will not light, and the error message sounds more serious than the problem is.

What the error looks like

Fit an untagged lamp into a VWD or DAD that is set to read tags. The module then reports UV lamp tag not ready, usually together with the ordinary UV lamp not ready status.

That is the detector reporting that it cannot find a chip. Nothing has failed, and the lamp is fine.

What the chip actually holds

Agilent puts an RFID tag on its newer UV lamps and on its flow cells. The tag carries the part number, the serial number, the accumulated burn hours and the number of ignitions. The detector reads all of that into Early Maintenance Feedback, the counter system inside ChemStation, OpenLab and Lab Advisor.

Two counters matter for a lamp: Accumulated UV Lamp On-Time and Number of UV Lamp Ignitions. With a tagged lamp those numbers stay with the lamp and are read automatically. That is the whole convenience of the chip.

Three outcomes, decided by one setting

The detector’s lamp-type setting decides what happens. The exact wording is different between software versions, driver versions and module firmware. Read the dialog on your own system rather than a screenshot from the internet.

flowchart TD
    A["Fit a lamp with no RFID tag"] --> B{"Detector lamp-type<br/>setting"}
    B -->|"Use lamp tag settings"| C["UV lamp tag not ready<br/>Lamp does not ignite"]
    B -->|"Manual lamp type,<br/>e.g. 5190-0917"| D["Lamp ignites<br/>and runs normally"]
    C --> E["Change the setting<br/>to a manual lamp type"]
    E --> D
    D --> F{"Heating mode<br/>correct?"}
    F -->|"No"| G["Error or no ignition.<br/>Try the other mode"]
    F -->|"Yes"| H["Running"]
    G --> F
    H --> I["Reset Accumulated UV Lamp On-Time<br/>in Lab Advisor EMF"]
    I --> J["Record the change<br/>in your maintenance log"]

The manual options are named after the part they replace. On the 8-pin detectors they come in pairs, one with continuous heating and one without. Only the right one will light the lamp. If the first one gives an error, use the other.

Some units also let you switch off a Lamp Tag required flag from Agilent Lab Advisor. That is reported by practitioners rather than documented for every module, so treat it as worth trying and not as a guarantee.

The step almost everyone forgets

An untagged lamp cannot store its own hours. So the counter does not reset by itself. It continues from the old lamp, and every hour you run is added to a total for a lamp you have already thrown away.

After fitting a chipless lamp, reset Accumulated UV Lamp On-Time in the EMF section of Agilent Lab Advisor. There is a reset button on the EMF page.

If you have the advanced version of Lab Advisor you can also set the limit on each counter yourself. The basic version uses Agilent’s default limits and will not let you change them.

If you skip the reset, the lamp still works perfectly. But you no longer know how much life is left in it, and that is exactly what the chip was doing for you.

Is this a problem in a regulated lab?

It is a change, and changes get documented. It is not a data integrity failure.

Nothing about the chromatography is different. The lamp is the same deuterium arc, the detector measures the same absorbance, and the method is not affected. What changes is that lamp-life tracking moves from automatic to manual.

So the honest position for a QC lab is this. A counter that filled itself in becomes a procedural step. A person performs it and records it. That is allowed, it is ordinary, and it needs to be written down: the setting you changed, the lamp you fitted, its serial number, and the date you reset the counter.

I would not recommend a chipless lamp for a lab whose SOP names the automatic counter as the control for lamp replacement. Change the part before the SOP, and you have created a deviation for the sake of a few hundred dollars.

Other vendors do this differently

The chip idea is not confined to Agilent, and it is not applied the same way.

Vendor	What carries a chip	What it stores
Agilent	UV lamps and flow cells	Part number, serial number, burn hours, ignition count, read into EMF
Waters	Columns, through eCord	Column name, part and serial number, injection count, first and last injection date, sample and sample set counts, maximum pressure and temperature with dates
Shimadzu	No chip found on replacement D2 lamps	Lamp hours tracked by the instrument or by the lab
Thermo	No chip found on replacement D2 lamps	Lamp hours tracked by the instrument or by the lab

Waters is the interesting comparison. eCord is attached to the column instead of the light source, and the history is stored in the Empower database, outside the projects. It answers a different question. Instead of how much life is left in a part, it records what the column had already done when a given result was produced.

For Shimadzu and Thermo lamps I could not find a chip-recognition scheme on the replacement lamps. If your instrument is one of those, the whole question does not arise and a chipless lamp is simply the lamp.

Before you order

Find out whether your detector reads a tag. If the instrument shows lamp hours without anyone typing them in, it does.
Check whether your SOP names that automatic counter. If it does, decide the SOP question before the purchase, not after.
Ask whether you have basic or advanced Lab Advisor, because it decides whether you can set your own counter limits.
Decide who resets the counter and where it gets recorded.

If the answer to step 2 is that the SOP depends on it, buy the chipped lamp. The saving is not worth the paperwork.

What I will do

Every lamp in the catalogue states whether it carries a chip, taken from the supplier’s list rather than assumed. Where a part is available both ways, say which one you want and I will quote that one.

If you are not sure which you need, send me the detector module number and tell me whether you work under GMP. That is enough for me to answer it.

Sources

From vendor documentation:

Agilent Community, New UV lamp not working, lamp tag not ready error. The error text, the manual lamp-type option, and the instruction to reset lamp hours in Lab Advisor for untagged lamps.
Agilent Community, Checking and modifying LC Early Maintenance Feedback counters. Accumulated UV Lamp On-Time, ignition count, the reset button, and basic against advanced Lab Advisor limits.
Agilent Community, Resetting LC UV lamp hours in Lab Advisor.
Agilent, Diode Array Detector WR and Multiple Wavelength Detector user manual. RFID tags on flow cells and UV lamps.
Waters, What information does the eCord Intelligent Chip Technology record and What is an eCord.

Not from vendor documentation, so treat it accordingly:

Working chromatographers on the Chromatography Forum describe the paired continuous-heating options and the Lamp Tag required toggle in Lab Advisor. They match what the Agilent articles say. But Agilent has not published them for every module.
I checked Shimadzu and Thermo replacement lamp listings for any chip-recognition scheme and found none. That is an absence of evidence rather than a confirmed no, so if your instrument is one of those, ask rather than assume.