So I would wire a cap to the ATTiny pin2 (or just a jumper wire from pin2 to pin6 to use the on board cap)?
And this uses stock off-time FW or the ports need changed?
Thx
FW needs mods. I was planning to work on them tonight but haven’t gotten around to it yet. PWM out must change pins, ADC changes are needed for LVP.
Yes, you could just wire up the cap as you described.
I’m not 100% positive that we can easily do the offtime cap & battery monitoring at the same time using the stock setup, but we’ll see once I wade into it. The offtime cap needs to be compared against a lower reference voltage than Vcc, so we normally compare it to the 1.1v internal ref. The battery monitoring voltage divider needs to get compared to Vcc. We’ll see how that pans out! I think it will be OK.
Oops, this is just me being crazy again! Totally insane. The ADC doesn’t return a boolean value (maybe the comparator does?). The code is already fine to do both. Ugh, the things I say sometimes. The only thing we need to tweak is voltage we compare against and all the values.
Note to self: divider gives 1.2v @ 6v input
Well, I’ve gotten mixed results from the first offtime firmware effort. I should have done it in stages, but I made all my changes at once. Oops.
I changed everything I mentioned before:
- moved PWM pin from 5 to 4
- added RUN_PIN on pin 3
- changed ADC reference from the internal 1.1v ref to Vcc (2.48v or 2.50v or something
) - changed LVP values
Please review my changes here.
Oops, I see a problem already - I didn’t remove the STAR2_PIN = PB0 define from the code. I’m not sure if that’s causing any problems, but we are using PB0 for something else. I can’t imagine that it’s doing anything good.
The firmware fires up and runs the driver at 100% duty cycle but does not successfully change modes at all, ever. I even shorted VCC to PB3 using my tweezers and then cycled power to the driver a couple of times. I could not get it to change modes. Hopefully this is due to my mistake with the PWM output pin?
C_K wrote:
Damn I ordered 6 of the v6’s literally 10 minutes ago!
Contact OSH Park and ask them to subsitute in v07. I've done it before. They were really friendly and happy to accommodate.Still should work or no?
Wight I can air wire a 13A in and start on the coding with you but if I order an LD-29 from fasttech I’ll be waiting a month+, I have the worst luck with their shipping every single time I use them. Do you have a spare I can buy from you for testing purposes so I don’t have to sit around waiting?
I went ahead and ordered 6 v07’s I’m not concerns about asking them to swap them out for $.80. It’s not worth Lean’s time even tho I know he’d do it so I’m not even gonna ask.
Unfortunately I’m not in much better shape than you. I only have 1 available to work on currently. And I just lifted a pad on that one! It still works, but I need to be more careful. If I was in better shape we’d get you hooked up. Be sure to spend the extra few cents on USPS First-Class. I always do on FT orders. For a single LD29 it’s $1.44, for two LD29’s it’s $0.88, and for 4x LD29’s it’s free anyway…
I know what you mean. When we’re talking about <$3 for PCBs I try to order as soon as I suspect the layout might be ready to go. If it turns out I’m wrong, well. … sad face: I just re-order. It’s worth it to me to speed up the time to have them in hand and possibly learn that changes need to be made. At least that’s what I keep telling myself. How’s your Eagle learning process going?
For the record when I was testing yesterday I noticed that the driver gets plenty hot. Regardless of whether the components can handle it, I’d pot the thing for sure.
In addition to what Cereal_killer said, note that OSHPark is totally rolling right now. From what I’ve seen they are doing 2-4 panels per day currently, so when you order you’d only have a tiny window to ask them to do manual corrections for you.
Was just trying to offer a friendly solution. C_K seemed upset about just ordering v06. No good deed goes . . .
That was just an FYI, not a reprimand! 
Did you notice what components got hot?
I was doing other stuff, so I did not pay close attention to that. I only worked on the bottom side. On the bottom side, it seemed to me that the FET and sense resistor both got hot. The sense resistor became hot faster I think.
:~ strange, both of the FETs’ are rated >8A on their datasheets.
I assume that it should’nt get really hot on ~2.8A ;\
[quote=HiTiT]
:~ strange, both of the FETs’ are rated >8A on their datasheets.
I assume that it should’nt get really hot on ~2.8A ;\/quote] I was working on measuring other stuff. Soon I’ll take a closer look at where the heat is really coming from.
I’d say that at 100% duty cycle this driver produces a good bit more than 2.8A out, although I have not measured. Probably around 3.5A+ output. The input at 8.2v (IIRC) was ~1.85A. That’s 15.17W input, so make whatever efficiency assumptions you want (80-90%) and figure the output based on that.
I dug this up just to post I found out what the black SMD component was, it’s when a SMD cap and resistor have a love child no seriously its a component called a ferrite bead.
Also my v06 boards are coming so I expect the 07’s to be a day two behind, as is usual with FT I don’t expect the driver till around middle to late August.
As far as I can tell an SMD ferrite bead would be a choke. I do not think that the component is a choke. If it was a choke I think resistance should be lower than 100k. Take a look at the DC resistance figures for some. They are ~100ohm or so and as I said in the OP this component is 100 k ohm.
My v07’s shipped yesterday!
A member here graciously sent me an extra LD-29. A good thing since mine is getting a little worse for the wear (from moving the PIC on and off, soldering wires to the pads, etc). Anyway it was ordered from FT on the 8th and arrived on the 17th with USPS - so 9 days from start to finish.
The new version LD29’s are somewhat different. I’m not yet sure what the implication of this is.
- still uses the PIC 12F683 MCU
- FET appears to be the same, an SO8 package 4435
- the diode is now SBM1040.
- an SO8 FET has been removed This missing part appears to be IR’s IRF7821.
It seems to me that IRF7821 was replaced with the tiny 8205S, but I have not looked into this closely.
SBM1040 is a 5A continuous diode in the Powermite 3 package. This package seems to be about the same size as PowerDi 5 but may be superior? I’m not sure that Powermite3 is actually superior to PowerDi5, but it’s definitely a modern high power diode so that’s good. The old diode was marked 3912 and I have no idea exactly what it was rated for. Maybe 3A, who knows. The new diode is certainly better!
The 4435 FET I assume to be a 30V P-Channel trench MOSFET. The new one is marked 4435AG and the old one was simply marked 4435.
Come to think of it, I do not see a buck controller on either the old board or the new board. There are no sanded components and nothing jumps out at me as a buck controller. Odd. I’ll have to follow the CSN trace and see where it goes from the sense resistor. I wonder if it’s going through a current sense amp and feeding directly to the PIC. If so, we need a whole new firmware 
Your new one’s from FT? My two shipped out yesterday ,I wonder which they’ll be.
I know about the good high res pic’s in the review, could you possible post a comparison shot of the two drivers physically sitting right next to each other orientated the same please?
I’m just wondering cause I’ll be ordering a PICkit3 this friday (I have 3 current projects centered around PIC MCU’s, I’ll probably be pretty much completely switching here as soon as I can and redesigning the 17dd for the 8kb PIC MCU of my choice) but is the only reason for all this to be able to program these with no additional equipment from the other AVR programming we all do? Is the MCU not able to be flashed (with the correct PIC equipment) cause its locked or something?
First of all: sorry Cereal_killer, I have no immediate plans for more photos. I’m really off of taking pictures right now. That’s why I’ve only been using my phone and only posting shots I deemed necessary. Also I don’t have a stock LD-29 available to put it next to.
That said, there’s really nothing you need to know that you can’t glean from what I wrote and the pictures I took.
I have no idea what the state of the PIC controller is.
My new one is from FT. I’m sure you’ll get the same thing, the listing on FT actually says that it’s the updated version.
I have not taken an in-depth look into the matter, but my current assumption is that the new version is just a simple refactor to make better use of space with the savings from going SO8 to SOT23-6 on that N-Channel FET.
May I ask what motivated your switch from Atmel to Microchip?