88/89 Inverter discovery

[KDirk]

Having recently changed a failing dash inverter (the one with the black case) on my 88, I decided to see if I could pin down why it failed (i.e. lights on the left and right switch pod would go out and return intermittently).

So, I removed the plastic casing from the inverter (pretty well destroying the case in the process) but got the circuit itself out intact. It is potted in a rubbery compound that can be pulled off carefully with a dental pick and small long nose pliers. Once I had most of the small circuit board exposed, I discovered two things:

1. The circuit board is riddled with cold (cracked) solder joints. The board itself is "raw", there is no conformal coating on the foil side of the board. This makes the failed solder joints easy to spot.

2. The inverter circuity is deceivingly simplistic. One transformer (voltage step up) 3 resistors and 1 Motorola TO-220 power transistor. That's it, and is so simple the circuit could be rebuilt for under $3 (for the resistors and transistor, the transformer - in theory - should never go bad) and some good re-soldering work.

The only other caveat is that a new enclosure will need to be devised to properly insulate the high voltage circuit against inadvertent contact since the original casing gets damaged in disassembling it.

I guess my point in posting this is that at some point good used spares will be impossible to find and this gives a path to repairing failed inverters. While A few of us here have stashed away some spare NOS pink (console) inverters, the black (dash) inverters have been unavailable as new parts for several years now.

I suspect, much as with the 90/91 IPC, most of the failures of the inverters can be traced to marginal soldering on the circuit board. The circuit is so simple, it is unlikely that any of the individual components would fail unless the output side of the inverter were shorted to ground for a prolonged period of time.

Not sure if this is of much interest right now since working spares can still be found, but I am always looking for solutions to problems with the electronics in the Reatta. I believe that is where much of our future challenges lie in keeping these cars on the road.

KDirk

[SeanR]

Great explanation. Why did they take DC voltage and change it to AC? Seems like such a wasted effort.

[KDirk]

Sean,

This was done because the backlighting in the two switch pods (headlamp pod on left, wiper pod on right) the gear shift shift indicator, the legends on the 6 "hard" buttons on the touchscreen and the mirror/window switches are all done with electroluminescent sheet, rather than regular light bulbs.

EL sheet requires high AC voltage to illuminate (a concept roughly similar to fluorescent lights). If regular light bulbs had been used they could have run directly off the 12V dc supply, but in order to achieve a certain "look" EL was used. This is what gives the backlit portions of the parts mentioned above their cool, even glow.

It also made the parts easier to design as the EL sheet is as thin as paper and can be cut into irregular shapes. Light bulbs would have required sockets and dispersion lenses to get the same effect. These both add bulkiness to the design of the parts. If the Reatta were made now, rather than 20+ years ago, I've no doubt they would have used micro surface mount LED's for this, but that technology wasn't ready in 1988.

KDirk

[Ronnie]

Kevin, Where is the voltage stepped up for the CRT? Does it use a flyback transformer like a TV?

[DAVES89]

Ronnie did you get that email I sent you on the led upgrade to the Reatta inverter?

It was from the original manufactuer of the inverters.

Edited June 11, 2012 by DAVES89 (see edit history)

[Guest moldymac]

What I have wondered about this inverter setup is that being it outputs 110v AC, couldn't one of these failed units be replaced with one of those cheap AC inverters they sell at most stores for powering a laptop or a cell charger in your car?

[SeanR]

One is pink and the other is black -so is the voltage output different?

Can't we just buy a group of these inverters? Thought I read in a post somewhere that the OEM still makes these but that a bulk order was required.

[DAVES89]

Both are discontinued. However the last time I did a group buy the pink [dash unit] was still in inventory. The black one [console unit] is out of stock and no longer available.

The companys name is Endicott Research

You can contact them to see what their order/shipping requirements are.

I just forwarded to Ronnie some info from Endicott about an LED upgrade option. Hopefully he will post that soon.

[Ronnie]

Here is what is currently in the vendor section of my website:

Endicott Research Group, Inc.

Vendor for 110v inverters that power the light pods. GM part numbers: 1642274 (black - center console lights) and 1642275 (pink - dashboard lights) Phone: 800-215-5866

[padgett]

One thing doesn't make sense: what is the transformer operating on ? The ripple voltage ? A transformer does not work on DC. YWTK

[KDirk]

Ok, to address the new questions that have come up:

1) The CRT operates on 12VDC input, it's high voltage is generated by a flyback transformer just as with any other CRT. This is completely independent of the EL backlighting inverters.

2) The inverters output 110VAC, but it is not 60HZ. I have not measured the frequency, but most standard EL operates best at about 400Hz. So, a regular inverter used to create a household outlet from 12V is not suitable. Also, these types of inverters are inefficient, noisy, and will not dim the EL lighting by the dimmer control on the dash.

3) As to Padgett's question, I am assuming the simple circuit creates pulsed DC at high frequency which is then stepped up with the transformer. It is a very (surprisingly so) simple circuit. I expected more than what I found in that little plastic box, that's for sure.

As to the difference between the pink/black units: I think the difference is output capacity. The inverter has to be designed to drive a certain surface area (usually spec'd in square inches, at least for the US) and so it is more an issue of current than voltage difference. That said, the pinout is the same for both. I have not bothered to try the pink one in place of the black (or vice versa) but I expect it would work, albeit with a difference in brightness.

Since the pink one runs the gear shift indicator, the mirror/window switches and the CRT hard button legends, it is probably a bit less powerful than the black one which drives the left and right switch pods. Keep in mind that in terms of square inches of surface area to be lighted, the left/right switch bank is more than the console switches, gear indicator and CRT surround, but not by much.

KDirk

Edited June 12, 2012 by KDirk (see edit history)

[Guest Mc_Reatta]

Kevin, need to correct a slight misconception you have with the CRT hard button lights. They are not electroluminescent strips nor are the console accessory switch lights above the gear shift. Those are traditional small light bulbs. So the console inverter only lights the gear shift indicator and the window and mirror switch labels, and the dash inverter only the two switch pods. Otherwise your comments are spot on.

The electroluminescent lights present a capacitive load to the inverter rather than a resistive load like a light bulb does. Brightness is a function of EL age and total area, and the voltage and frequency of the AC supply.

Looking at the web site that Ronnie posted for the manufacturer, they still make EL inverters that could be used if we figure out the area of the EL strips they power, but they seem to have discontinued the plastic case style that we have in our cars so some reconfiguring would be required.

After having gone thru the joy of replacing the dash inverter before, I have a plan in the back of my mind to do away with the EL strips and inverter, and replace with LEDs to remove the reliance on an an obsolete part the is that also so hard to get at. Believe with the colored LEDs available now, the appearance of the lighting would be very close to what the ELs provide, or allow one to use multicolor LEDs to designate the different functions on the switches to modernize the look of the dash.

[DAVES89]

McReatta glad you have an '89. That means what you are working on is what I will one day need!

[KDirk]

MC-

Not to be argumentative, but I stand by the statement that the frame lighting on the CRT (legends for the 6 hard buttons) and the window/mirror switch are backlit by EL. I know as I have had all 3 parts completely disassembled. The CRT frame has 2 very slim EL strips (about a 1/4" tall by 6" long) to backlight the hard button legends. There is a very small amber LED (not quite surface mount, more like 1/2 the size of a T-1 LED with conventional wire leads) as well for each of the six switches. The switches themselves are of the common metal snap dome over circuit trace variety used prior to the advent of surface mount micro-tact switches.

The mirror and window switches have a small EL sheet overlaid by a diffuser (translucent white piece of plastic) and that is then covered over with the decal, a precision cut piece of anodized foil on the early version switches, and a lithographed "foil look" decal on the later replacement parts.

The only switches on the IPC and console that are lit by conventional 12V micro lamps are the four in the IPC itself (trip reset, fuel gage, E/M and test) and the console switch bank with the fog lamp, trunk/glovebox/fuel access, and lamp retract switches.

Your explanation of the inductive vs. resistive load is correct, and most helpful.

KDirk

[Guest Mc_Reatta]

Not sure of the 88, but the 89 FSM definitely shows the AC from the console inverter only going to the gearshift, window and mirror switches, and the 12 volt dimming signal only, going to the CRT (thru pin A4) and console switch lamps.

So either we have discovered yet another error in the FSM, or there is an inverter built into the CRT somewhere that supplies the AC voltage to those frame strips you describe.

Also note that the variable DC voltage that powers both of the inverters is daisy chained through each of the components that will receive the EL AC voltage before reaching the inverter to turn it on. This leads me to believe that running the inverter without all of the EL loads attached would be detrimental to the EL strip and / or the inverter itself, so they built in these interlocks to prevent the inverter from operating without all the switches connected.

[KDirk]

Hmm,

As far as 88 vs. 89, the components and harness for the EL lighting and the respective wiring should be the same. Both years used the same mirror/window switches, gear shift indicator, CRT and headlamp/wiper switches. There are a multitude of differences between the two years, but in this particular regard they are virtually identical.

I hadn't looked that closely at the CRT pinouts in the FSM. I had only gone at it from the standpoint of dis-assembling the actual components for cleaning/repair work. I have a CRT torn down on the bench presently, I will inspect the "side car" circuit board, as if there is a separate inverter for the CRT frame EL, it would almost have to be on that board. The other circuit board (beneath the picture tube) is all driver and power supply circuitry for the CRT display itself (i.e. horizontal oscillator, flyback xformer, focus block etc.).

Interesting note on the interlock as well. I had not made note of that, but have not had a need to look at the EL driver circuits that closely as I have not had to do any troubleshooting in that area (yet) aside from the need to replace a failed dash inverter.

Will check the CRT interface board and report back in the next day or so with my findings.

KDirk

[KDirk]

Ok,

Looked at the 88 FSM, and the CRT does not have an input for the EL high voltage lines from either inverter. The FSM circuit diagram is also a bit misleading in that it shows all the components with EL backlighting represented by the standard light bulb schematic symbol, when they are not incandescent lamps. I guess there was no better way to represent it on the schematic, but I can see where that would leave one to think the back lighting was done with bulbs and not EL. Especially since the console switch bank does use lamps, and this is indicated with the same light bulb symbol on the circuit diagram.

Now then, the question is how the EL backlighting in the CRT frame is driven, since it is not coming from either of the external inverters. The interface (side mounted) circuit board does not have any inverter circuitry, but the main circuit board has a few components clustered together that look to be a rudimentary circuit much like that contained in both the pink and black inverters [i.e. a small transformer and a few discrete components]. This stands to reason since the power input and other power supply for the CRT are on the main circuit board. It also stands to reason that the CRT has it's own EL inverter since the Kent-Moore CRT test tool will illuminate the hard button legends when the CRT has no connection to the car. So, the EL driver high voltage has to be generated within the CRT unit.

The downside of this is that if in-board inverter on the CRT dies, the hard button legends go out. The only way then to get them to work again is to rebuild the inverter circuit within the CRT. I have never seen a post regarding the hard button legends burning out here on the forum, so perhaps the CRT's internal inverter is more reliable than the two outboard ones. In any case, this is another oddity that will probably become an issue the older these cars get.

KDirk

[Guest Mc_Reatta]

Well Kevin, I think we uncovered a secret that wasn't realized before.

I can forgive the use of the lamp symbol in the diagrams due to the fact that in 88 I don't think there was widespread use of a symbol for a LED yet no less a symbol for a light emitting capacitor which is what the EL panels really are.

But the lack of the AC being delivered to the CRT and the use of the lamp symbol there as well as the IPC and console switches did lead me to believe that mini lamps provide the background lighting for the touch screen hard switches. I have not had the need to do more than remove the frame and touch screen from my CRTs as a unit yet.

Since the CRT is a fairly highly unreliable module in our cars (but for reasons not associated with the backlighting function to my awareness yet either), there should be a source of those inverter circuit components available from cannibalization even if new replacement components aren't identified.

Yet another reason to hand on to old inoperative CRTs as at some point in time, like IPCs, rebuilding will be the only option.

[KDirk]

Yet another reason to hand on to old inoperative CRTs as at some point in time, like IPCs, rebuilding will be the only option.

MC-

I am a big advocate of saving unique assemblies (like the CRT, IPC etc.) even if they are not working for this very reason. Those of us lucky enough to be able to keep these cars viable long term will find, at some point, that working used parts are simply not obtainable any longer. Those, like myself, who have a bit of an obsession with this car would do well to have our own modest stockpile of these types of parts for as long as we own the car. Other than the space required, there is no harm in doing so.

I imagine that will start being a problem somewhere around 10 years from now as all of the marginal vehicles that were left, and most of what are presently average condition daily drivers will have long since been scrapped by then. The occurrence of these cars in a salvage yard will get to be very slim after another decade of "culling the herd".

That will leave the exceptionally well maintained and show quality/collector car owners to battle over what scraps are left. I don't relish the thought of that since it will drive up parts prices, even if the car itself doesn't gain a big collector following; but that is yet another discussion.

Getting back on point, we are already at the stage where good repair practice dictates a failed CRT unit should have it's main circuit board completely rebuilt. All discrete and passive components (resistors, capacitors, diodes and probably most transistors) should be removed and replaced with new parts. This solves the common failure modes; loss of raster, horizontal oscillator and deflection function etc.

Since the in-board inverter appears [from my cursory examination] to consist of a small transformer and a handful of other discretes/passives just as with the two out-board modular units, replacing all of these parts except for the transformer in a CRT unit rebuild should ensure the continued operation of the EL lighting in the CRT frame. At least until the point of the EL itself failing, and then we get to figure out how to replace that too.

KDirk