Ericomp thermal cyclers

Lawrence P. Casson lpcasson at phoenix.princeton.edu
Sun Jan 16 12:32:23 EST 1994


In article <2h1g6i$dpj at news.u.washington.edu>, Tim Buss
<tbuss at fred.fhcrc.org> wrote:

> We have an Ericomp 'PowerBlock' thermal cycler.
> ...
> IT WASN'T MY IDEA TO BUY IT!!!!
> Anyway, we have had many problems with this machine. It has already been
> repaired once but has now had a second nervous breakdown! Our loan
> machine has now gone on strike in sympathy.
> I would like to hear from you if you have been having problems with one
> of these machines.
> Does it work?? Are you happy with it??

We would also appreciate any comments regarding reliability of thermal
cyclers.

Some of the following comments regarding the demos we have had might be
erroneous due to lack of time to use the machines.

Please note that our Ericomp demo was an EasyCycler TwinBlock and not a
PowerBlock.  TwinBlocks are water cooled, so they don't go below ambient
temperature without a water chiller.  This may be important if you use a
high temperature polymerase with 3'-5' exonuclease activity (proofreading).
 I've never had a problem with reactions using Taq polymerase that were
left at room temperature overnight.   I would guess they don't use Peltier
devices for heating, since they don't use them for cooling.  I don't know
why our lab evaluated a TwinBlock.  It could be that PowerBlocks (which use
Peltier devices) are not available with a 96-well microtiter format.
PowerBlocks contain two blocks and are less expensive than TwinBlocks (see
below).

1) Programming

Ericomp TwinBlock - Obtuse programming.  They really could have done this
alot better.  You need to first enter a cycle, and then incorporate it into
a program.  In a lab with multiple users, it becomes harder to keep track
of standard programs, and the requirement for two programming processes
introduces greater room for error.  It seems that the only way to use it as
a constant
temperature heat block (Sequenase or Taq sequencing of ssDNA template) is
to
enter a cycle and a program.  All of this is much easier on both the MJ
Research machine and our old Coy machines.  Both of these have a mode for
direct control of the temperature without resorting to a program.

MJ Research - Easy to program.  My only complaint is relatively minor.
A new program may be inserted at any point in the list of programs which
already exist.  If your lab keeps a list of programs contained in the
machine, the order of the programs in the machine will change relative to
your list unless new programs are added specifically at the end of the
list.  It would be nicer for compulsive people like me if programs could be
assigned a specific number
(eg. 1-99) in addition to (or instead of) a name.  As stated above, there
is a mode for direct control of the temperature.

2) Temp. variation

Both machines worked fine for a somewhat testy assymetric PCR reaction that
I use to prepare template for sequencing.  Using 96 well plates, there
seemed to be no variation that could be blamed on variation across the
block.  The lack of product in a few wells was more likely due to poor
starting materials.  I have become a convert to 96-well plates and thin
wall tubes for these reactions.  They seem to allow better results with
higher reliability and shorter cycling times than the standard 0.5ml tubes
that our old Coy uses.

3) Cost

Hot Bonnet ($995) is a heated lid option for the MJ Research PTC-100
($3995).
The Ericomp EasyCycler Twinblock TCX20A ($5390) has an Evaporation Control
Device (ECD) accessory ($695).  As stated above, a twin block PowerBlock
TCX-40 costs less ($3295).  We had a quote from USA/Scientific for the
Ericomp TCX20A for approximately the same price as the MJ Research w/ Hot
Bonnet, but it's unclear to me whether the ECDs are included (the demo had
them).  One selling point for the Ericomp is that it has two spearately
programmable blocks, and the $695 is actually for two ECD's.

4) Reliability

Ericomp: 3 year warranty
MJ Research: 2 year warranty

Ericomp: water cooling, meaning solenoids, I assume.  Hey, they fail in
washing machines where the number of cycles per day pretty low.  Also, why
does it require the use of a temperature probe?  A probe has no effect on
even temperature distribution across the block.  If you need to optimize
your reaction, you have to do it for machine and tube wall thickness,
whether a probe exists or not.  How long before someone breaks it off?  (No
cycling without it I assume)  It takes up a well in a 96-well dish which
means 23 sequences instead of 24 (very minor inconvenience).
MJ Research:  Can anyone comment on the reliability of Peltier devices?
The one on our old Coy failed twice, but that was a relatively inexpensive
machine that was purchased 5 years ago.  I was impressed by how little heat
the MJ Research machine pumped out.  It must pump heat in and out of the
heat sink, and the cooling fan isn't called into service all that much.
Pretty amazing compared to the old Coy.

5) Miscellaneous Remarks

To me, heated lids seem overrated.  They are useless with 96-well plates.
While ten 96-well plates are more expensive than 1000 common 0.5ml
centrifuge tubes, this is probably not the case with the thin wall tubes.
And you can easily cut up the 96-well plates if you only need a small
proportion of the wells.  I think the annoyance of lids and individual
markings is far greater than the bother of a drop of mineral oil added to
each tube.

6) Impressions

Ericomp (EasyCycler): microprocessor controlled low-tech in a package that
looks high tech - I don't have a problem with low-tech since it's usually
reliable.  MJ Research: concealed higher-tech.  Light weight, doesn't seem
to produce much heat, easy to program.  Will the Peltier devices last?

7) Preference

Given the choice between the watercooled Ericomp TwinBlock and the MJ
Research PTC-100, I'd lean towards the PTC-100 if I was signing the
requisition, especially if I knew more about the relative reliability of
the two machines.


Now, can anyone help out with the initial reliability question?

-Larry



More information about the Methods mailing list