Replacement Ignition Capacitors

It’s now generally recognised that the quality of the internal bonding and construction of replacement distributor capacitors is lamentable. A supplier of replacement capacitors has gone to some trouble to commission capacitors made to the original Lucas specification. Unfortunately, not the best of places to start!

Lucas capacitors used beryllium-copper wavy washers to spring load connections to the capacitor element, whilst the replacements use a “bee-hive” spring of zinc plated steel to make its electrical connections (see photo 1).

Photo 1 – Internal construction of distributor capacitors.

Both approaches fall well short of industrial standards and introduce significant series resistance; in the replacement sample that John James obtained, the series resistance measured 3.65 ohms. Whilst initially this is not high enough to affect the capacitor’s function, it is significant in that it predisposes the capacitor to failure in much the same way as the old Lucas capacitors.

The spring loaded electrical connections deteriorate with time, possibly due to localised heating from the high pulse currents, which would tarnish the surface. As the series resistance increases, so too do the very conditions that cause the connections to degrade. This represents a “runaway” situation, leading to series resistances up to 100 ohms and above.

As previously mentioned in the article on ignition systems (Issue 31, August 2015), the increasing series resistance shortens the contact-breaker life and then eventually the ignition fails due to insufficient energy at the spark plug to ignite the fuel-air charge.

Possibly, the only real improvement in the new replacement capacitors over some previous replacements, is the use of an epoxy-like potting compound to seal the wire-link end. Some previous replacements relied on a “rubber bung” to both seal and tension the electrical connections to the capacitor element. The types of connections detailed above are in stark contrast to the plasma welded connection of the recommended pulse capacitors (see photo 2).

Photo 2 – Pulse rated capacitor with plasma welded connection.

A meaningful test for these capacitors would measure their series resistance, but as this requires special equipment, its relevance seems to have been understated. The construction techniques used are so misconceived as to defy comprehension in this advanced age. Is someone taking the originality issue too far?

The case of the replacement capacitor was zinc plated steel, but traces of corrosion existed inside, not an issue with the Lucas unit which was made of tin plated brass, so some quality aspects of the Lucas unit are unique.

Eric Worpe

Ed’s note: The replacement sample which I obtained and sent on to Eric was purchased from Martin Jay, who is better known as ‘The Distributor Doctor’.

I thought it only fair to forward Eric’s article to Martin for his comments; these are reproduced below preceded by the text of a covering e-mail from him.

Text of an e-mail sent by Martin Jay to John James dated 28^th April, 2016.

Thank you for the opportunity to see the article on industrial capacitors, which I am familiar with. Please see the attached document in response to concerns over our products.

I would like to summarise in saying that the original Lucas specification & method of construction, which is the same as ours, has stood the test of time.

Only the relatively recent inferior copies over the last 15 years or so have given the capacitor an unreliable reputation.

We make all our capacitors to the same exacting standards and have sold over twelve thousand with just two reported failures.

I need say no more.

“Attached document” follows………

Replacement Ignition Capacitors / plasma welding technique.

With hindsight, it looks good to criticise the old Lucas capacitor or any near replacement, but how long have they served the industry well?

OK if new techniques come along like Plasma welded connections that can only be a good move forward. Plasma welding of the connecting wire to a capacitor has now been a standard in the wire-ended types (as shown in Photo 2, which is not an ignition capacitor) destined for printed circuit use for some time, but the ignition capacitor is a little bit different.

One of the main requirements of a conventional ignition capacitor is the means to keep a good intact connection between the foil element and the case housing allowing for expansion caused by heat between the foil capacitor element and the case. These connections must be flexible necessitating; the need for the internal connection springs used to keep the connection between the foil element and the case. Hence the Beryllium / copper wavy washer used by Lucas. As long as a replacement capacitor uses a spring, or any means of keeping the connection intact that will not corrode, then the result is the same. The same applies to the case that Lucas used which was a brass tin plated case, but a replacement steel zinc plated case is OK as long as it does not corrode. Another major criterion in the Lucas design is the fact that the foil element must be a loose fit in the case so that the spring methods used for keeping the connections at the lowest possible resistance can be effective.

If the new plasma welding technique can be done on a flexible braid to connect the case to the foil element and the foil element to the insulated wire end, the allowance for expansion is not so critical because the plasma welded flexible braid takes the place of the conventional spring designed to keep the connection as low as possible. The same applies to the insulated wire end if this can be connected direct to the opposite end of the foil element then a spring connection here is not required meaning a better product. The foil element, in my opinion, must still be a loose fit in the case to allow for longitudinal and lateral expansion.

Another improvement must be an epoxy seal on the insulated wire end instead of a ‘rubber bung’, but I would have thought no one building replacement capacitors relies on a rubber seal to keep out damp and corrosion!

In conclusion.

1) The old design has stood the test of time for many years and cannot be so poor that a panic change should be implemented.

2) If a plasma weld can be used on an ignition capacitor (not as shown in photo 2 a wire ended type) then this will remove many elements such as the spring-loaded connections, which can only improve the product.

3) If an epoxy seal can also be used to replace the ‘rubber bung’ then a good product can only result.

Ed’s further note:

I copied Martin’s e-mail and attachment (his document) to Eric.

Eric has come back, saying that continuing the discussion may only serve to confuse the issue even more, but makes three main points, which are:

1. a primitive spring-loaded connection is unsuitable when a reliably low resistance capacitor is needed
   
2. traces of corrosion found on the inside of the zinc plated steel case leaves a question mark about quality control
   
3. there doesn’t seem to be any sound reason why a modern capacitor such as the one suggested could not be used as they are heavy pulse duty rated to 85 degrees centigrade and the very low series resistance precludes internal heating.
   

Eric has indicated that he is willing to write an article on how to fabricate replacement capacitor modules.