Page 2 of 2

Re: hip9011 integration

Posted: Wed Jul 15, 2015 4:11 pm
by AndreyB
I just need a tiny little break from this painful testing, but I am planning to retest different hardware setups soon.

Re: hip9011 - crystal issue

Posted: Wed Jul 15, 2015 6:29 pm
by Tomin
kb1gtt: 470p capacitor is used for connection from external source of clock (CPU or ...) into HIP/TPIC in the PDF = without a crystal.
But you use it in a setup with crystal. It is totally another situation.
I think you should not tight crystal circuit with 470p to "ground", because it will be hard to start oscilation due to unappropriate loading of circuit.

Don't know anything specific about Secu in practice with knock sens., sorry.
I think one of a developer of the SECU is here, is not ?

Re: hip9011 - crystal issue

Posted: Thu Jul 16, 2015 1:44 am
by kb1gtt
I'm not sure about SECU developers. Perhaps there are some here. It's cool to see that they have made good progress.

I agree that C169 should not be populated typically.

One common trend I've noticed on these other layout that don't seem to have noted problems is that they seem to use the thru hole style XTAL instead of the SMT style. I wonder if that could some how be the difference. Perhaps the longer thru hole leads add some thermal mass and slow down the heat in the XTAL and some how prevents damage to the XTAL, or perhaps the thru hole makes some difference in capacitance / inductance that's beneficial. I do not see why thru hole would be better, but it does seem to be a potential common trend. We are using a ATS08A-E for the STM. Perhaps we should also combine the components for commonality purposes. Looks like it would fit physically.

Re: hip9011 - crystal issue

Posted: Mon Oct 22, 2018 4:51 pm
by kb1gtt
Was originally at the below link. It make more sense in the hardware section.
https://rusefi.com/forum/viewtopic.php?f=5&t=778&start=89
Are there any reports of this working reliably? I see 6MHz XTAL vs our 8MHz, perhaps that has proven to start reliably, or perhaps it's just as buggered as ours. I recall our no-start issues were on something like 10% to 20% of the boards. They have a resistor in a slightly different location, which I recall we had in a prior revision, but it proved to be non-important. So I don't think the reliability is any better because of this slight different in a resistor placement. I suspect the magic is 6MHz instead of 8MHz.