Hello Folks
After having lurked around for a while, I am new to this forum. I adopted an 81 Double Six with a pre HE engine. One of the problems it has, is a intermittent failing rh cylinder Bank. It seems to be one of the 4 driver transistors in the ECU DAC 1936E. They are older RCA 17343A NPN types which I cannot find an exact substitute/replacement for. Before sending the ECU in for repair or replacement I would give changing out the transistor(s) a try and would really appreciate any advice where to find the right parts.
Best regards
Gerd
Thank you Robin. No success there.
Best
Gerd
**
I’m not too sure if an exact substitute is required, Gerd - the ECU is a pretty crude device…
Frank
xj6 85 Sov Europe (UK/NZ)
**
Thank you for answering Frank
I tried BD243C. That is a universal BJT NPN Transistor. As far as I see it should have the same pinout. It did not work.
Best
Gerd
Gerd, there was some relevant info in a 2003 thread on the XJS forum. Do a search for this topic
[xj-s] 3 ECUs and a car that doesn't run-electronics pundits?.
Might be of some use as they discuss the 6CU
Best regards,
Jim
**
That a replacement part does not change anything, Gerd - may just mean the original part is OK and the fault is elsewhere. Or the replacement part has the same fault…
The sole purpose of the ECU is to ground the injectors once every engine revolution - using sensors to compute the duration of the grounding…
The finer details of ECU operation is partly based on development - but crudely; all computations leads to selecting injection duration from a 1288 bit ROM. The result is of course used for all injectors…
The separation is 4 groups of 3 injectors - each group with its own driver transistor, each of which through a power resistor, initially bypassed at triggering. I mention this for the record - you probably know it all…
However, you don’t mention the type of trouble with the ‘right’, (A), bank? Is it confined to a group of injectors; 1-3-5 and 2-4-6 (from the front)? And have you conclusively eliminated the resistor pack as a factor?
Or is the problem with the A-bank of a different kind…?
Frank
xj6 85 Sov Europe (UK/NZ)
**
Thank you Jim and Frank for the answers.
Jim: I fount the mentioned thread before when searching the forum for that transistor type but got no answer out of it for my problem.
Frank: I checked the resistor, the wiring and the ecu for loose contacts, measured the injector pulses with an oscilloscope. The rh bank pulses have the same peak voltage but they are way shorter than on the lh bank. It seems that there are two signals generated in the ecu and that the injector groups are split behind the transistors. Tha aj6 engineering description linked in this forum show how the power output works. When I replaced the transistor with a BD type, the lh bank changed from running bad to not running any more. Maybe I can find a uses ecu I can part out or…
Best
Gerd
**
I assume that the ECU computes injection duration, Gerd - then feed the result to all the power transistors. Which simply grounds all the injectors for the prescribed time.
Mind you, the detailed internals of the ECU is beyond comprehension, but there is no way it can compute each bank separately - no separate inputs for each bank…
However; as the ECU triggers injection; the initial pulse bypasses the resistors - delivering 12V directly to the ECU for rapid injector reaction. (Actually; power may go both past and through the resistors). The ECU then (presumably) switches off the direct path - just switch between the two…?
This explains the peak voltage similarity between the banks - and the loss of ‘sustain’ power through the resistors on the ‘right’ bank. How the ECU does this, I don’t know. But as two transistors serves each bank and all ‘right’ injectors are equally affected - it cannot be a single transistor fault…?
Have you actually checked pulses at each injector, or at least one injector in each group of three with the 'scope - to identify the relevant power transistor. The transistor are bespoke for each group of three injectors. Using the 'skope at the resistor packs connections should show the pulse width identical for each group, of course - but would be different (but equal between groups) on each side of the resistors…
My point is rather that if the engine misbehaves - the anomalies may also relate to other factors than the injectors…?
I’m bewildered; you say the ‘left’ bank is misbehaving - and quits when replacing the transistor. Which also implies that that transistor works the ‘left’ bank - while the injector signal anomaly is on the ‘right’ bank…?
Frank
xj6 85 Sov Europe (UK/NZ)
**