Victors are known to be a low frequency controller. IIRC, Theyre supposed to be 120Hz.

The OSMC/IBC ran at about 2.5Hkz which was pretty high for the designs in that Era. Vantecs were about 700Hz I think.

The wave you posted does look a little weird, but Im not familair with the output drive part of Victors, do they have a snubber or filtering after the fets ?

It looks sort of like a very short "Full Off" time, which is required with some gate-drive configurations to allow the gate-drive charge pump time to build up its voltage. although Victors dont use the Hip chips that the OSMC do, so Im not sure how they boost their top-side bridge fet drive voltages.

Do they use the same N-Type Fets all round the bridge like OSMC designs, are are they a p + N type Fet bridge with different Fets on the top and bottom legs ?

Might be worth scoping the Gate line on the Fets to see what the drive signal looks like, rather than the output with the motor inductance , bridge design and other factors complicating the waveshape.
_________________
Great minds discuss ideas. Average minds discuss events. Small minds discuss people

Hi Brett, thanks for the feedback! There doesn't appear to be any snubber components; no large inductors and just the one input filter cap. All 12 FETs are the same IRF3703. As soon as I find one of the Spin controllers, I will get the case off and look at the gate drive, it should prove interesting.

I had a sneaky idea with the spin controllers: they are the same PCB layout but the motor is wired differently. The output terminals are bridged and the motor is wired between the bridged terminals and the battery negative. I think the ESC must be switching the top half of both bridges together and not switching the lower FETs at all. Does that sound at all plausible? If that is the case, I can save time & money by just replacing the upper FETs. If the 885 ESC was getting warm using three FETs per leg, the spin controller using six FETs in parallel, each with half the RDS, should have no problems at all.

Desoldering FETs:

Problem solved! I changed the soldering iron tip for a smaller one so I could see where it was touching more clearly and changed the desoldering tip for a larger one with better flow. Using tools both at once got enough heat into the board to suck all the solder out and I have half the old FETs off the controller.

New parts:

Its raining parts again; the replacement FETs, heavier duty fuses and a pair of LIPO packs just arrived:



The LIPO packs will be made into a 6S center-tapped pack for Shiny. Its a bit early to know for sure, but the A-spec Turnigy batteries seem to be as good as the big name brands. With all this stuff arriving, its going to be a very busy weekend!
_________________
Australian 2015 Featherweight champion
UK 2016 Gladiator champion

Quick update: the gate waveform is a classic PWM although its maybe a bit low at 8V - I guess the Victor's strange output will remain a mystery. Its time to move on to the FET upgrade and some testing. Even if the new FETs don't work out for weapon drive, they will still be better than ever for driving the wheel motors.
_________________
Australian 2015 Featherweight champion
UK 2016 Gladiator champion

Magic smoke

I tried a spin-up test for Scissorhands, using the hacked TZ85 from last year:



Everything went well at low speed, with the motor running quieter due to the higher switching frequency. When I increased the speed the motor responded until I reached close to 100%, when the motor speed suddenly dropped off, which is what happened at last year's Nats event. Backing off to around 90% got the motor spinning properly again and then about 20 seconds later the ESC exploded and released more magic smoke than I have ever seen from electronics before ! It filled up the workshop and was so thick that I couldn't see the power switch to turn the test off.

That's it for plan B; now I have to get that upgraded Victor going . Its hard to explain why the TZ85 exploded; its capable of supplying the 40A that the motor draws, its rated for more than 22V and there is only one way to wire it and have the motor start at all so it wasn't a wiring issue.

In better news, Scissorhands is 30% reassembled and has a few reliability improvements.
_________________
Australian 2015 Featherweight champion
UK 2016 Gladiator champion

damn…that sucks …i guess the video camera wasn't rolling either right?
_________________
Miles Blow - Julie Pitts
www.mulesfilm.com.au
www.wombokforest.com.au

-Pickasso- Vivid Sportsman champion 2015

Unfortunately no - it would have made an excellent "don't try this at home" clip.
_________________
Australian 2015 Featherweight champion
UK 2016 Gladiator champion

FET forest felled:

All the FETs have been de-soldered, along with the smoothing capacitor, which was on its last legs with metal fatigue (pun intended):



The next step is to replace just one FET on each leg of the H bridge to test if the rest of the ESC is functioning. After lying around since 2008 and getting very toasty during the de-soldering, the board could be dead. If the Victor works with just 4 FETs installed, I will add the rest and give it a test on the Mag motor - now know as the moroe of doom after killing so many controllers .
_________________
Australian 2015 Featherweight champion
UK 2016 Gladiator champion

Hey Nick, is the smoothing capacitor the one in the middle? And how essential is it to the controller's operation?

Electronics isn't my strong point, and that capacitor broke off of one of my 883s. I didn't know what it did and it didn't seem to affect its performance so I've never looked at fixing it, but I might try and sort it if it would be better in the long term.
_________________
Jamie
RogueTwo Robots
www.rogue-two-robots.vze.com

With great power, comes great reliability

How'd you go about extracting the board from the case? I'm not sure if they are all like that but the two 885s i have seem to be potted in around the edges with a really strong epoxy. Wouldn't mind making some new cases for them at some stage

The A-spec should be good. Think i may have linked a big comparison test someone did on rcgroups but they are at the least equal to the better cells. Definitely better than the thunderpowers. Demon was using the standard nanotech last year after i puffed the g6. Not bad at all for the money
_________________
www.demon50s.com - Minimoto parts
http://www.youtube.com/user/HyzerGlen - Videoooozzz

It just an educated guess - the smoothing cap is probably preventing the high current PWM output from interfering with the supply to the CPU and FET driver circuits. That could crash the CPU or cause the FETs to not fully turn on, which then leads to over-heating. As Glen discovered with a different ESC, its not always necessary but its better to have it 'just in case'. If you get around to replacing the cap, it needs to be a low ESR type for best effect. The one I have is 1,000uF, 35V.

None of the Victors I have disassembled have had epoxy; they have a an odd protective coating that looks a bit like jelly and might be a very soft urethane. I found that running a razor blade between the case and the PCB eventually released the parts - after quite a fight. I have several cracked or broken cases, so any plans for a 3D printed replacement would be most interesting.
_________________
Australian 2015 Featherweight champion
UK 2016 Gladiator champion

The cap is to absorb the voltage spike from the battery when the PWM turns the fets off.

The inductance in the battery and its wires is a bit like water running through a pipe, when you turn the fet off its like slamming shut the tap. The pressure on the tap keeps rising and you get the water hammer. The only difference is with a FET you get fire shooting out rather than water. The cap is like having a hydraulic accumulator on the tap, it gives that moving water some place to go absorbing the pressure spike.

You want to replace that cap with a low ESR cap with the highest ripple current rating you can find. The IBC had a great cap on it, that's why it melted its legs off when paired with the crappy Chinese brushless controllers.

As you let the voltage rise higher you need less capacitance so the more modern fets/controllers are getting away with crappier caps. They also don't seem to mind if they explode from time to time ;->.

I'd suggest perhaps putting a dob of silicon (or better something that doesn't need air to cure) under the cap when you replace it for mechanical holding down type stuff. But something you can rip apart when its time to replace it.

So in short I would be really careful running it without the cap, anything is better than nothing and the lower the current draw the less the voltage spike will be.
There will be some low value probably ceramic caps on there to bypass the logic supply. Yeah looks like there may be tantalums on there (the big orange things) be really careful with them they will explode like a mofo if subjected to abuse in terms of voltage.
_________________
Mechanical engineers build weapons, civil engineers build targets

Its alive!

The upgraded Victor actually works after lying on the junk pile for six years !



The test is with only four FETS, so now I will add the remaining eight FETs and try the ESC with the Mag motor - let's hope it does better than the TZ85. Jaycar dissappointed me (as usual) with the low ESR cap, so I found a slightly larger value one in my spare parts bank - not sure if its a low ESR type though.
_________________
Australian 2015 Featherweight champion
UK 2016 Gladiator champion

Anything is better than nothing.
When you start to load it up put your scope probes on the battery terminals and see what its doing.
Fire your thermal camera at it and keep an eye on the cap.
(its a better way than trying to do it with maths in my book, its mostly just guessing otherwise)

Other than that, looking good man, nice work,
You will probably get different looking results when you have the inductance of a motor hanging off it but that should be like 2X as good in terms of heat waste as it was before.
_________________
Mechanical engineers build weapons, civil engineers build targets

Thanks! I was not expecting too much from an old ESC; these Victors are tough little suckers. I tracked down the capacitor specs using Digi-key; its a Panasonic part with a ripple current rating of 3.3A and an impedance of 0.017 ohms @ 100KHz. No ESR is given but the range is described as 'low impedance', so its probably going to work OK.
_________________
Australian 2015 Featherweight champion
UK 2016 Gladiator champion

All sorted

I ran a test on the upgraded Victor and the new FETs have made a difference:



After two minutes at full power the FETs stabilised at around 32 Deg C, which is the best result of any of the controllers I have used to date. Its a mystery why the low-side FETs at the front get hotter than the high-side counterparts but at these temperatures I don't really care.

If I can find the spin controller and swap the FETs, the results should be even better.
_________________
Australian 2015 Featherweight champion
UK 2016 Gladiator champion