direct replacement to a LM1785T

You are always bragging that you are smarter than everyone else on the sci.electronics.* newsgroups. Now is the time to put up, or shut up. Then your co workers can go back to shoving you down the steps.

Was the circuit and surrounding environment examined for complete physical integrity before repairs were attempted?

Was this the only change in the unit made to restore function?

If these devices are in a bridge arrangement, was the survivor of the pair retained/identified?

Were any/all pull-outs saved for soft probing and examination by the vendor or other agencies?

What has been done to attempt to induce failure in your standard test units? Will the units run continually into a load short/open/grounded? What are the assembly's demonstrated operating environmental limits under normal functional stresses?

A '5 ohm across the rails' is not a very useful description of node impedances in any circuit. Without a schematic, other descriptions are equally so.

In answer to your question, devices like the venerable L149 and L165 (from pre-1984) have the same package and pin-out. There have to have been a host of others since then.

RL

We simply replaced it with a new one.. no other changes.

The other unit in the bridge of the same part was/is fine.

Hmm, well since I was the one the originally designed the circuit, I really don't know by whom I would be presenting this to?

As I stated before, I designed in 25 Ohm R's on the output, you can short the output via the R to common with no problems. I tried to make it idiot proof..

And to make a another note I forgot about, the output didn't short to any rails, only the Vcc and -Vee shorted together with no sign of short showing any where else on the chip.

I don't know if that means anything..

Excuse me, that was a LM1875T my mistake..

If a forensic study is inconclusive, presenting the failed part to the mfr (Nat Semi - after determining valid origin and provenance), along with all relevent paper from the study, is the lowest-cost method of determining what internal damage is present in the sample.

How inductive is the load?

What happens when the input(s) are open-circuit or faulted otherwise? Having to maintain a virtual ground, or driving one end of the bridge when the other has slammed permanently to a rail under a signal fault can produce stress that is unanticipated.

The mfr will be the one most likely to see relevance in the failure mode.

How is power established in the circuit? What happens when a rail is removed temporarily, arrives late, or sticks around too long at power-down - simulate specific connector pin intermittency where possible. Check connected assemblies, connectors and power sources from the original failure situation.

It's the only flavor of lm1875 they sell, so the t is irrelevant.

RL

Look at the subject line where he called it a LM1785T.

--

aioe.org, Goggle Groups, and Web TV users must request to be white listed, or I will not see your messages.

If you have broadband, your ISP may have a NNTP news server included in your account:

There are two kinds of people on this earth: The crazy, and the insane. The first sign of insanity is denying that you're crazy.

Ok, the load is inductive but I have TVS diodes with fast blow links on the output. The fuse links are on the output side so that the TVS will clamp and open the links. The links at that section were fine.

I intentionally caused a large induced voltage at the output leads originally just to test the abilities of the TVS diodes and links with a HV scope attached on the sensitive side. The diodes clamped nicely with no sign of a pulse above the diode saturation level so i'm confident that isn't the issue.

The bridge outputs are joined via 25 ohm R's and then into the load. Even if they slammed on in any manner between the R's and protection clamps I have, I don't see how that is irrelevant ?

If either one of the rails should fail to initiate, it wouldn't matter,

The power supply is a basic bridge rectifier as a full wave with a CT xformer as the common. Also one thing to note. Both LM1875T's are on the same supply rail. I would think if there is some psu issue, wouldn't both have a problem ?

In case I didn't state it before, I'm using a +/- 15 volt supply and operating the bridge from the + & - side. with this config, I can either get a ~ +/- 14 volts out to the load as a single output with respect to common as a bipolar out or use both for a floating bridge which gives me +/- 30 volts.

Ok, I'll take a looky to see how they would look for alternate replacements.

AT the moment this happens to be a bring it home on the weekend project and I got a test load on it now performing random output. everything is looking fine at this point.

Thanks.

How inductive is the load? Please answer in units of measurement.

What range of inductance is expected?

So you haven't done this? Doing it is probably quicker than fiddling around. Fool with input signals, in operation, where the signal is not carried by a direct, hardwired, component-free conductor.

If the connection between the rails is a direct, hardwired, component-free connection, then simulating a fault is a waste if time. If theres a connector, solder joint, or component that can go wrong, it's best to check out the effect, if you haven't already done so.

This is basic practise to gain experience in accurately predicting open/short component fault effects, in the future. It also increases your understanding of actual circuit performance, allowing more reasoned handling of field trouble in the future.

When you say that you 'can' get somethig, do you mean that this is what is actually done? If so, include variations in your failure-mode-forcing physical trials on 'healthy' test subjects.

One of the things you should try to record and control carefully is your test environment. This required so that others may duplicate your results without actually having to visit your home. Keep a list of the equipment (serial numbers) used as well as procedures and test methods, accurately dated.

A camera is becoming a cheap and useful tool for recording failure sites, test set-ups, dut physical environment, assembly and dut appearances etc. This is something that can be done at the failure site without excessive fiddling or interference with the actual sample, giving the locals something to do, and giving you something to look at and possibly think about.

You should try to insist that field failures be initially bagged and sealed before return. The bag shouldn't be opened, subsequently, by anyone unwilling or unable to perform a complete failure analysis.

RL