DFGT Circuit board help

  • Thread starter Thread starter luizsilveira
  • 71 comments
  • 26,342 views
L

luizsilveira

35
United Kingdom
United Kingdom
So I thought about creating a new thread since I've identified the problem is not the ffb motor but the circuit board. Old thread here.

Just a recap: I have two faulty DFGT wheels. One doesn't turn on at all when plugged in. The other one only blinks the GT light when plugged in. All buttons work on both but no steering.

I tested the voltage to the FFB motor: first wheel has 1V, second wheel has zero.

I've been testing the PCB components and comparing the readings. Mostly testing the board without current and on diode mode. They are all the same but one guy. Reference my board again. The black component all the way to the right, I don't know what that is. Likely a diode? It would make sense if those are faulty since that plug right next to it is the ffb motor plug.

So, reading it on the board (since I'm just comparing I thought this would be enough) and diode mode:

Board #1 (no power): 655 one direction, ~510 to the other. Busted diode? I'm unsure since it's a bidirectional so I don't know if it should read anything to the other side.

Board #2 (GT light blinks): blinks 1460, then open circuit (no reading) on both directions. Likely a busted diode?

Component reads 9NP6KE27CA, and the best match I could find is this (without the first 9).

Looks like this could be the reason why wheel #2 is not getting any power to the ffb motor. Unfortunately I have no soldering iron at hand or I'd test swapping them right away. But... in any case, if those are diodes and if the very little I know about diodes is right, then they are both busted (in different ways).
 
Last edited:
That is aTVS diode. (Transient Voltage Suppression). What do you get when you check the resistance (ohms) from one side of the diode to the other? If it's low, it's at least part of your problem. - It's there to short out and sacrifice itself if the voltage goes too high to protect other components. It normally protects against voltage spikes of very short duration when the switching power supply s witches of and on very rapidly (PWM).
 
Last edited:
Thanks. That's roughly what I found through the part number, and now I'm trying to find where to buy them in Edinburgh. Best I could do was 5 of them for £2,50 via ebay. But the thing is: if only one of them is busted it makes no sense for me to buy more (since that would mean something else is bad in the other board). I'm contemplating buying a soldering iron and just switching them "to see what happens", although the expenses to get this thing working are adding up. If it'll ever work :P

Right, resistance readings. I'm a bit lost with the scale as it seems to return different values with the 20k scale and the 2000 scale. Let me put them below:

Board #1: 20k scale reads 2.16 one way and 2.10 the other. 2000 scale reads 523 one way and 671 the other.

Board #2: 20k scale reads infinite both ways. 2000 scale quickly blinks something between 1000~1800 and then reads infinite;

If I'm not too dumb, #2 seems like a busted component. Question is: is #1 good? Should I take it out and put on #2 board?

Thanks a bunch eKretz, your help has been invaluable!
 
Don't use the 20k range. Most meters like that have a 200 ohm range - use that one. You should always use the smallest range possible for the thing you're trying to check. What do you get? I would say board 2 sounds more like it's working, but not sure. The diode is supposed to sense a transient voltage, then clamp itself into an effective 'short' to keep the voltage spike down, then release again, ready to monitor for another. It sounds like #2 is doing that. I don't have any spares around or I'd check one for you.
 
eKretz
Don't use the 20k range. Most meters like that have a 200 ohm range - use that one. You should always use the smallest range possible for the thing you're trying to check. What do you get?
Click to expand...

200 scale: Board #1 blinks crazy numbers (different each time) then goes to infinite. Board #2 stays on infinite (doesn't even blink).
 
Hmm. Neither is shorted then. Maybe they need to be checked out of circuit. Let me take another look at that data sheet. I can check them, but it sounds like you probably don't have the equipment. I would desolder them and hook them up to a DC power supply with current limited to a low value like 20mA, then bring up the voltage to it's clamping voltage in the data sheet and see if it shorts. Without a DC power supply it's tough though.
 
Aye, indeed I couldn't do that. I missed the shop times today to buy a soldering iron and pump. Should have bought them when I had a chance. But while on the one hand this is kind of fun, on the other hand I feel like I'm being victim of sunken costs. Damn Fanatec wheels that are out of stock right now...

Anyway. I'll probably buy the soldering stuff tomorrow, then I can test them off-circuit. But the DC power supply with varying Voltage I won't have access to I'm afraid. And no one in town seems to be willing to look at and troubleshoot a board.
 
OK can you possibly plug the wall wart into the board and power it up and see if you have 24V before the TVS? Just keep clear of the line voltage!

Also - not many places will troubleshoot electrical devices anymore. The people with the skill to do the troubleshooting and the time it takes are too expensive for them to be able to make any money. Most would rather just buy new stuff than pay almost as much to fix the old. Most of the stuff that gets repaired at the component level these days is either VERY expensive gear or people that do it themselves.
 
Thanks again! Yeah, I remember repair shops who would do this kind of work all the time - some 20 years ago :) TVs, VCRs, all that fun stuff. Not any more apparently.

Yes, sure I can plug it in and test it. But what do you mean by "keep clear of the line voltage"?

I did test the ffb motor plug itself. Board #1 had 1V and board #2 had 0V. But that's obviously after the TVS.

By the way, I don't think I'd know what's "before" and what's "after" the TVS. But testing both sides shouldn't be a problem, right? My only doubt would be: where should I put the ground probe?

Cheers.
 
I mean since you have a bare board be extremely careful not to touch anything with your fingers or the probes of your voltmeter that is connected to line voltage (you guys use 240VAC right?) Just the standard safety warning.

Check the board where the red dots are:



If there's a coating over the board there (I can't tell from the photo) probe the legs of the ICs that are connected to the copper colored ground planes indicated by the red dots. You might check it again after you desolder the TVS if it's inconclusive to see if that is the problem.
 
Ah, ok, thank you. Yeah, I thought you might have meant that with the "line" but I thought it better to check it out. Sorry but English is not my first language so I was afraid to lose something in translation. Yeah, I think it is indeed either 220 or 240V here. Most appliances are bivolt nowadays so I don't even remember checking it in the past few years.

I apologize but the two red dots might have gotten me a bit confused. Should I test one probe into each red dot, or the black probe into each red dot while the red probe is touching the TVS's legs? :confused: As far as I know there shouldn't be a problem unless I touch two or more components with the same probe, right? I hope that's right, but as you can see I don't know all that much...

Ah, ok, now that I read it over a few times I'm fairly sure you mean the red dots as neutral, so it shouldn't make a different if I use either one? I'll test it out and report back.

I don't even have words to thank you enough at this point!
 
Updating with the readings:

Board #1: between 0.85 and 0.98V on the top leg of the TVS. 0V on the bottom leg;

Board #2: 0V on both legs.

One more interesting thing: every time I fire up board #1, the power adaptor makes a high pitch whiz. It happens with both adaptors (I have two). There is no whiz when I use the adaptors to power either board #2 or my old MOMO. Very strange.
 
One probe in each red dot. And never mind about the line voltage... I forgot your wall wart takes care of the conversion to DC so you shouldn't be in danger of a zapping anyway.

The high frequency noise could be because there's either a current draw or because there isn't one. Some supplies make noise when they're working, some when they're idling. If that TVS is clamped it may cause the problem but it's possible it might have damaged other things too. The numbers from when you tested it are kind of inconclusive, since they vary.
 
Ok duh... Wasn't thinking on that one. Did you get a soldering iron to remove the TVS yet? If it's clamped it will draw that voltage down but it should be reading as a short if that's the case. You may have to just start following traces back. Another thing you can try is just to power up both boards by plugging them in to the wall and checking voltages in different places - comparing them to find anything that isn't the same on both boards.
 
I did think about that but I didn't do it in the end. Comparing voltages all over the board I mean. I wouldn't know the places to check I think. I only checked the board unplugged and with diode mode. Everything reads the same but that TVS.

I didn't get the soldering iron yet. Maybe I'll get them over ebay as it's cheaper there but it takes a couple of days to arrive. I'm not sure I'm super inclined to buy even more stuff if it's at best a very long shot though. :S
 
Right but some components will check funny in circuit when they aren't energized. If you remove the TVS and find you have the proper voltage then you are good to go - just make sure to replace it. If not you can follow the traces back to see if something doesn't have the voltage that you expect, or you see a difference/discrepancy. If you want to start checking some voltages just stick your negative probe on a board ground plane (Or even the negative pin where the power comes in should do) and start poking the positive probe in different spots following the traces back from the motor plug. Compare board to board right next to each other - since you have 2 power supplies it should go quick, and it won't cost you anything but time.

If you are thinking of scrapping the wheel you should at the very least snip out the TVS using wire cutters and see if it's the problem first.
 
Last edited:
luizsilveira
One more interesting thing: every time I fire up board #1, the power adaptor makes a high pitch whiz. It happens with both adaptors (I have two). There is no whiz when I use the adaptors to power either board #2 or my old MOMO. Very strange.
Click to expand...

I broke the earth pin off of my old Driving Force Pro's power supply by standing on it (luckily for my foot it was sharp side down), the other day I found it, cracked the casing off of it, rehoused it and added a power cable so I could at least gain a 24V 0.75A supply for use with something else, and yeah; it whines like mad when energised and whines even more when you give it a load, I suspect it's just coil whine from the transformers. My iPhone does it too, and it's one of those things where once you hear it you can't ever ignore it, for me anyway.

Also if you give up on trying to fix one or both of these boards I'll happily buy one or both from you, I'm only down the other end of the island from you so postage shouldn't be too much of a hassle. I was probing my Driving Force Pro's board but then I put it down for five minutes, looked back and it was gone, now I can't find it anywhere so I'd like some replacements without buying a whole wheel. I'm also intrigued to have a look at the circuit board that's inside the steering wheel itself... Just don't throw them away!

Good luck with your project, I don't think I can contribute anything more than @eKretz has so I won't even try. It'd be interesting to know what the FET voltages are like, though.
 
eKretz
[...]
If you are thinking of scrapping the wheel you should at the very least snip out the TVS using wire cutters and see if it's the problem first.
Click to expand...

Ah, thanks, I may try something like that. Although I can imagine it'd be a bit of a pain to solder it afterwards if the legs were cut. I'll probably just get a cheap iron+pump, they should sell for some £5 on ebay.

This is a question I had actually: should the ffb motor get energy (i.e., work) if I remove that TVS completely?

neema_t
I broke the earth pin off of my old Driving Force Pro's power supply by standing on it (luckily for my foot it was sharp side down), the other day I found it, cracked the casing off of it, rehoused it and added a power cable so I could at least gain a 24V 0.75A supply for use with something else, and yeah; it whines like mad when energised and whines even more when you give it a load, I suspect it's just coil whine from the transformers. My iPhone does it too, and it's one of those things where once you hear it you can't ever ignore it, for me anyway.

Also if you give up on trying to fix one or both of these boards I'll happily buy one or both from you, I'm only down the other end of the island from you so postage shouldn't be too much of a hassle.

Good luck with your project, I don't think I can contribute anything more than @eKretz has so I won't even try. It'd be interesting to know what the FET voltages are like, though.
Click to expand...

Thanks a lot for your help! The funny thing with the whining adaptor though is that both of them whine but only if attached to this specific wheel. Neither of them whine if I use them with any of the other wheels. So my (quite amateur) guess is that there is something with the current that is being pulled from that specific board - either too much or maybe even too little?

As for buying a wheel from me, even if I do manage to fix one of them I'll certainly still have the other. So one way or another I'll have a spare DFGT. Just pop me a PM with an offer and we can see it from there. I've tested and they have perfect ffb motors and optic sensors, so they should be worth something as those are common things to fry and brake in these wheels. I also have a black MOMO with a dodgy calibration problem fwiw.

Lastly, yeah, I've seen this engineer fixing a G25 by replacing the MOSFETs. Apparently the are prone to go boom as well. But I had a good look at them and man... I wouldn't have the soldering skills to get them straight even if I was sure that was the problem. So I didn't even bother too much with it. I might give them a voltage test when I get to removing the TVSs.
 
After you snip the TVS diode it is pretty trivial to remove the legs actually. You just grab the leg with a pair of tweezers or small needle nose pliers and heat the joint and the leg will pull right out. If you aren't very well versed with soldering here's a tip: before attempting to desolder a joint, put a small dab of fresh solder on the tip of your iron - it will help with heat transfer so you can get the joint melted faster (by increasing the surface area in contact with the joint when the liquid solder conforms to the joint). Sometimes snipping a component out is the best way to do it when there's minimal access space - so it won't cause any desoldering problems for you if you snip it.

The wheel should work without the TVS diode, but you shouldn't run it in a race or anything without it. Just a quick test should be OK though. Its purpose is to block very high, very short duration voltage spikes from the power supply or motor.
 
Ah, thanks. Good tips.

What I meant is: if I snip it and the TVS diode is actually good it won't be so easy to reuse it really will it? Or (just occurring to me now) should I just snip right where it joins the board so I can re-use it just with shorter legs?

Edit: yeah, no, just had a look and, ulike all the resistors, the TVS is already touching the board so there is no way I can cut part of its legs and still re-use it easily. I think.
 
Last edited:
Thanks eKretz! As I posted in luiz's old thread, I too thought I had motor failure and did some quick testing and threw the wheel aside for a while, thinking I would just have to get a new one.

Tonight I had a few minutes (hours) and decided to give it a go. I recalled this thread from before and thanks to your help I have determined my issue appears to be a TVS that is stuck closed. I was testing/verifying the 24dc from the cord first (ok) and next from the back of the jack when I plugged in. There I only had 3 vdc! I knew something was shorted/failed/grounding so after I checked a few places I came here and realized this is very possibly the answer.
I took out my iron, took it out, plugged back in and calibration cycle starts (and i now have my 24v back)!

My next step is to order a new one. I will post on my results after replacement. I would like to post also a quick guide with pics to test for this and fix for it (if I can find some time, lol, I still have to do this for an encoder wheel fix I posted a while back). I believe this will be very valuable for those who have experienced a similar "failure", and don't realize what a simple / cheap fix it is.

Many thanks again!
 
Hey that's great Lenny, good job! I'm glad to have helped! Make sure to get that diode replaced before you do any racing - it's there for a reason and protects some delicate components down the line when the voltage is rapidly switched.
 
Hmm, sounds like you may have a failed transistor or FET. Pop the TVS diode in and see what happens but that's my guess.
 
Get some good clear photos of the board if the issue persists after the diode change so we can follow the traces and read the component markings and we may yet be able to find the culprit. Do you have a diode test function on your multimeter?
 
Well the diode is already out and that is what is happening.
I do have that function on my multimeter, I am fortunate to have a nice Fluke due to my line of work. I am familiar with the basic concepts of electronics, and am learning more all the time. I am very grateful that you are sharing your expertise as I enjoy learning as much as (potentially) solving the problem itself.

I have the new Diode ordered and will update when I get it installed and we can go from there (I will try to post a pic of the board soon for you to look at in the meantime).
 
Sounds good, in the meantime I would check all the SOT23 transistors scattered around the board right next to the drive FETs to look for anything funny using the diode test function on your Fluke. Forgive me if you already know this, but put your negative lead on the single leg side and check the pair of legs on the opposite side one at a time with the positive lead. Usually you'll get a similar voltage drop on both. Some will be opposite polarity (NPN vs PNP) so you'll need to switch the leads.

If nothing odd manifests itself there, power the board and check voltage on all the pins of the driver FETs and report back. Just be careful not to short any pins with your leads! A simple way to do this a little easier if you have alligator clip leads is to clamp the alligator clip on a stick pin or needle and probe with that.
 
Just got my new TVS diode(s) yesterday. Question is how do I tell which way to install? Which side goes to V+ and which goes to V- ? There is some writing/labeling but nothing stands out as a marker for one side vs the other.

Edit: OK I did some wiki-googling, and now know there should be some kind of mark indicating the cathode. Also learned that TVS diodes are installed reverse-biased, so that means I connect my cathode to V+ and my anode to V- correct?
 
Last edited:

Latest Posts

Back