Hi Deborah,
Firstly, it is not unusual to have differences between set and actual values, certainly of 50V so this is unlikely to be the problem.
Temperature variation does sound to be your problem. The idea of a 'flat topped ratio peak' is so that these variations can take place without affecting your results. The drift correction in the software will also take care of some of these variations.
The problem is that with extreme variations the magnet properties can change slightly causing a shift of the magnetic field and hence HT. This is more extreme on some systems than others. If the HT drifts so much that it 'falls off' the edge of the flat top peak then no amount of drift correction will help.
The problem is obviously worse with sulphur because the flat top peak is so narrow.
As others have said, the room temperature has to be kept as stable as possible, hot or cold.
Cold is preferable because the electronics prefer to run cool and extremes of heat will cause faults or fast noise in the electronics particularly the vfc card.
The first tip I can give you is to run small batches and check/set the HT before each new batch.
Secondly, observe the temperature changes over the day and the effect on the HT. If the HT increases as the temperature increases then if you are starting a batch as the day warms up, position the HT on the left side of the flat top peak. As the temperature warms up the Ht will then have to travel all the way across the flat top before eventually falling off the peak, instead of just half way across.
Other advice is make sure the magnet is out of direct sunlight and make sure no draughts blow over the instrument. If users have air conditioning this should not blow directly on to the instrument.
Hope this helps,
Chris Pullan
Sercon Limited

>  from:    Deborah Rupprecht <[log in to unmask]>
>  date:    Wed, 26 Jun 2002 09:52:50
>  to:      [log in to unmask]
>  subject: Re: [ISOGEOCHEM] ANCA SL/2020 and room temperature
>
> Hello to all,
>
> I am working with ANCASL/2020. As we noticed last year during sulphur
> analysis, there is a influence of room temperature (more than 2 degrees
> variation is bad for the voltage regulator and it can not keep the HT set
> conditions).
>  Some days ago it was sunny and hot outside ( air from outside is pumped in
> the labs wothout cooling!!) and we were analysing nitrogen-- it was a
> mistake. During 120 samples we ´ve got increasing negative delta values of
> standard material (worst case -150 delta instead of 20). Peak scan after HT
> scanning line of 4046V was not placed in the middle of the peak but right
> before the peak lines are scaling down. "New" middle of the peak was at
> 4032V . But it did not help (new source settings were saved, of course):
> results of new batch run coursed weight dependence (weight/%N f standard
> material).
> Now I want to know:
> - is there somebody outside in the world-wide-MSlab who has made same
> observations?
> - why do results becoming worse even if the HT is optimized (middle of the
> peak plateau)?
> - who knows at which room temperature range (and humidity) an ANCA SL/2020
> is working well?
> - Comparing setting value and actual value of HT there is a deviation of
> about 50V. Is that usual? ( I once tried to set the HT at the actual valuee
> but is was getting worse).
>
> Thanks for any advice in advance,
> Deborah (Rupprecht)
>
>
> *****************************************
> *                                       *
> * Deborah Rupprecht                     *
> * Institut für Pflanzenernährung        *
> * (vormals Agrikulturchemie)            *
> * Rhein. Friedrich-Wilhelms-Universität *
> * Karlrobert-Kreiten-Str. 13            *
> * 53115 Bonn                            *
> * Germany                               *
> *                                       *
> * Tel. +49 -(0)228 -73 -2145            *
> *                                       *
> *****************************************