Updating HF welding generators by replacing

expensive obsolete tubes with modern types

By Dr. P. H. Foreman F.I.E.E.

INTRODUCTION
e2v technologies is increasingly being asked for help with equipment conversions to replace obsolete and expensive
tubes with modern rugged industrial types. The cost of conversion is invariably lower than the cost of purchasing a
direct replacement tube and, when carried out with e2v technologies’ advice, is generally straightforward. The benefit
of the conversion may not only be elimination of concerns about availability and lower tube replacement costs;
enhancing the performance of the equipment is often possible, because modern tubes are more efficient. Furthermore,
depending on the power supply, uprating the welder may be possible.
Many of our customers carry out successful conversions based solely on a set of recommended electrical conditions for
the replacement tube. However, if e2v technologies are provided with a small amount of extra information, we can help
to minimise downtime during the conversion, ensure optimal performance and frequently reduce the number of
additional components required.
The conversion may be divided into three parts: Mechanical, Electrical and Checking Conditions. These will be
discussed in turn.

Mechanical
Normally in sets with obsolete tubes plenty of space is available to accommodate the necessary changes. Additionally,
the proposed replacement is usually smaller. However, to avoid introducing parasitics, the original layout should be
followed as far as possible.
Modern tube anode cooling arrangements are more efficient, so the water flow characteristics may be different.
Typically, they require less flow, but at a higher pressure. This can normally be accommodated by adjusting the flow
trips.

Electrical
The most common requirement is for a new filament transformer, but often a tap change will give the required
adjustment. Modern tubes usually require less filament power so that even if fewer volts are required the current rating
of the transformer is not a problem.
The grid resistance may need modifying to a lower value with a higher current rating, but this has not been a problem in
conversions made so far.
Modern tubes take less drive at higher grid currents. The feedback may have to be reduced. Some older tubes ran at
high HT voltage and low current, but they were normally in the lower power ranges used for induction hardening rather
than pipe welding.

Checking Conditions
This involves recording the new operating values and comparing them with those recommended by e2v technologies. If
any significant differences are noted then we can check them on our computer programme and recommend minor
adjustments.

ACTUAL CONVERSION
These points will be illustrated by describing a recent conversion in Turkey. We are extremely grateful to Mr. Ural
Tamur of Bosas Profil A/S, Trabson for giving permission to use his experience and for providing before and after
photographs to illustrate the various points.
The set is a Nippon-Denko type FRT-200H designed to use two type 9T64 tubes.
Photograph 1 shows the equipment as it was working with the original tubes fitted. It can be seen that plenty of room is
available and the layout is uncluttered.
The BW1184J2Fs were held up in the cabinet to give an estimate of the pillar height to support them in a way that
changed the layout of the set as little as possible. Our distributor had already produced substantial (10 mm) aluminium
plates with suitable holes to mount the tubes. These were in fact cut down and the off-cuts used to remount the anode
blocking capacitors.

e2v technologies limited, Waterhouse Lane, Chelmsford, Essex CM1 2QU England Telephone: +44 (0)1245 493493 Facsimile: +44 (0)1245 492492
e-mail: enquiries@e2vtechnologies.com Internet: www.e2vtechnologies.com Holding Company: e2v holdings limited
e2v technologies inc. 4 Westchester Plaza, PO Box 1482, Elmsford, NY10523-1482 USA Telephone: (914) 592-6050 Facsimile: (914) 592-5148
e-mail: enquiries@e2vtechnologies.us

# e2v technologies limited 2003 A1A-TPD167 Issue 4, April 2003

527/5874
Photograph 2 shows the BW1184J2Fs in position. On the site, we had made a total of eight insulated pillars; the back
four were shorter to fit above the anode blocker support plates, which were in turn supported by four sets of the
original red pillars. In addition, we had made four adaptors to braze into the tube water unions to take the plastic water
pipe.
We had recommended the BW1184J2F which has integral braided filament leads to avoid maintenance personnel
making high current connections to the tube, because if they are not tightened properly the contact will arc with
disastrous results. The original cathode return was through low inductance straps from the filament posts and we were
concerned about these being replaced by higher inductance braided leads. Fortunately, this did not cause problems,
but another time we would recommend the standard tube using filament connectors with straps to reduce the risk of
parasitics and loss of efficiency. In most sets the filament, which is the cathode, should be as close to RF earth as
possible. Normally one side of the filament is earthed so the situation can be corrected by replacing one braided lead.
However, here the filament transformer is centre-tapped to earth so two straps would have been needed, making the
integral leads redundant.
Naturally a water leak was found when we first switched the water on!
The set was first turned on with no modifications at low HT. When no faults were found the full HT was applied.
The readings appeared reasonable, although we had expected to reduce the feedback, so the welding line was started.
To the untrained eye the weld was satisfactory, but the operator complained that the inner bead was uneven and the
weld split on the cone test.
It was thought that the problem was due to unbalanced loading of the tubes and considerable time was wasted
changing grid resistor values on the two tubes. It was later established that they were like that to start with and the
difference was almost certainly due to meter error! None of the changes affected the weld quality and we concluded
that the weld stitching was associated with too much feedback for the BW1184J2F. The excessive feedback was
causing grid currents that were too large for the grid capacitance to sustain a continuous oscillation. This resulted in a
condition known as a blocking oscillation, in which the oscillation occurs intermittently.

The circuit is a Colpitts and the feedback capacitor arrangement is shown diagrammatically in Fig. 1 and the actual
layout in Photograph 3. All the capacitors are 1500 pF.

C3

C5

Fig. 1 Diagrammatic representation of the feedback arrangement

The feedback is given by C3/(C3+C5) = 14.3%

TPD167, page 2 # e2v technologies

We chose to add a capacitor in series with C3 to give a ratio of 10%, see Photograph 4, but we could equally have
added a capacitor in parallel with C5. The feedback would then have been 11.1%.
The set was switched on and the same anode current was achieved with much less grid current, but more important
the weld was perfect! Having diagnosed the problem, we considered using the alternative feedback arrangement given
above to give more headroom against a blocking oscillation. However, when the set welded perfectly with the largest
pipe, we left the feedback as initially modified.
We also carried out a second conversion on a Nippon-Denko set, type NF-250. Here we experienced problems with
parasitics, because while fitting the BW1184J2Fs we had distorted the anti-parasitic suppressor chokes. In Photograph
4 the anti-parasitic suppressors are seen to be very substantial, whereas in the second set the chokes were wound with
3 mm copper wire. Once they had been reshaped to be uniformly wound along the resistor, the problem went away.

LESSONS TO BE LEARNED
Several practical lessons can be learned from the conversion described:
1. The scope of the work is well within the capabilities of a competent maintenance team, because all the conversion
work was carried out by the staff at the plant.
2. Prepare metalwork and necessary fixings in advance to minimise downtime. The whole installation took a day largely
because everything was fabricated as the work progressed.
3. Note the readings before converting to avoid wasting time trying to balance currents that were not balanced in the
first place!
4. If the feedback needs changing, do it straight away, do not wait for a bad weld. If the grid current is larger, ensure
that the grid capacitance is increased in proportion to the change of current.
5. Avoid radical changes in circuit layout and distorting components, especially the anti-parasitic suppressors.

CONCLUSIONS
Significant financial and operational benefits may be gained from making conversions to modern e2v technologies
triodes, using personnel already employed at the plant.
e2v technologies recommend conversions not just to replace an expensive obsolete type for lower replacement costs,
but also to reduce stockholding. Several of our customers have standardised on their highest power level and run the
tubes in the lower power sets at reduced filament voltage to increase their life, and have more than compensated for
the extra cost.
There is no need to be afraid to convert. At the low frequencies used in pipe welding, differences in tube characteristics
are much less important than in other applications. This means that very little risk is involved in converting equipments.

APPENDIX
Information required by e2v technologies to give most help in carrying out a conversion
a) Present range of operating conditions Va, Ia, Ig, Rg
b) Circuit diagram of the RF section giving present tube type
c) Photograph or sketch of the cabinet layout
d) Feedback details, i.e. actual components, values and ratings and sketch of layout details
e) Grid resistor details, i.e. actual components, values and ratings and sketch of layout details
This information will enable us to advise on layout, feedback and grid resistor changes as well as giving expected
operating conditions, which will increase the confidence of the people doing the work.

# e2v technologies TPD167, page 3

 Photograph 1 Original cabinet layout

TPD167, page 4 # e2v technologies

 Photograph 2 Converted cabinet layout

# e2v technologies TPD167, page 5

 Photograph 3 Original feedback capacitor arrangement

Photograph 4 New feedback arrangement with an extra capacitor (arrowed) in place

Whilst e2v technologies has taken care to ensure the accuracy of the information contained herein it accepts no responsibility for the consequences of any use
thereof and also reserves the right to change the specification of goods without notice. e2v technologies accepts no liability beyond that set out in its standard
conditions of sale in respect of infringement of third party patents arising from the use of tubes or other devices in accordance with information contained herein.

TPD167, page 6 Printed in England # e2v technologies