Mosfets?

[slider69]

Alright u guys...and MICROAMPS in particular if he is reading this...there has to be a way to mosfet the ZZ se. I looked at the PCB on my friend's, and there is a coating (looks like the black epoxy used to cover Mosfet mods on Microamp's cars) That coating is not to keep u from modding..It is to prevent short circuiting with the lights' PCB inside the body. But anyway, the coat should be removable, and the thing has to use those H-bridges at some point so....Anyone MOSFETTED one of these or got any ideas on that?

[crazydave]

First off mosfets are not a H-bridge. The H-bridge is the circuit that switches between foward and reverse in Bits, and the original Zip Zap. It is composed of NPN, and PNP transistors. The mosfets that we use are N-channel, so they can be directly swapped for the NPN trannies, and the have to be flipped to match polarities with the PNP.

Since the SE is digital proportional, there is no H-bridge, and I don't see any N-type transistors that seem related to to the drive.

I've already started scratching off the black stuff, it's just paint, chips right off, and at this point, as far as I can tell, that big black block right by the motor is what is controller the throttle. I deduced that because the connections to it are labeled forward and reverse. (see pic)

http://members.cox.net/strollerpusher/sepcb.jpg

So the mosfets we've been using are not going to work, and until someone breaks down this circuit, and finds a compatible component, with better performance, mosfets ain't happening in these. Bummer too, because I'm used to my FET'ed cars now, and these seem really wimpy to me.

[slider69]

Yeh...Im sorry i was unclear about the H-bridge...I did know it was the type of circuit. OH and i replaced the pnp transistors w pnp mosfets on my ZZ. But look at the front of the pcb where the black epoxy is on the PCB....There are transistors on the board but i guess they're for steering. I cant see the numbers on them..could there be transistors as a preamp for that large box?

[crazydave]

Originally posted by slider69
..could there be transistors as a preamp for that large box? That's what I'm wondering.

Since this is my only SE, I'm in no hurry to tear it apart. When I get another I'll get more aggresive about figuring out what i can do to the circuit.

[slider69]

BTW, What are u using to take off the epoxy, and does it just peel of or is it stuck to the components pretty well? Oh, and what is the number on the part...maybe i could cross reference it with a manufacturer and find out exactly what it is. U could try to wire the mosfets between the box and motor (like u would wire a regular transistor as an amp) couldn't u. i dont think that would cause u to lose the propo control.....

[crazydave]

The paint chips right off, I'm taking it off with my finger nail

I can't read the part number on the box, because it came off with the paint.

Adding a mosfet where it don't belong wont do anything. Mosfets increase power because they have 1/10th the resistance of a transistor, and allow more current to flow, but they do have resistance, so if you're adding it, and not replacing something with it, it would only add resistance. Besides, it's a switching device, what would you be switching?

[slider69]

On the contrary, most transistors can be wired as an amp if i am not mistaken....it usually involves wiring base to collector so the transistor immediately turns "on" when voltage is applied. And mosfets do amplify by other means than resistance change, otherwise "stacking" would be useless, right?

[crazydave]

A transistor or mosfet peform the same function. In an amp the transistor is just turning the signal rapidly off and on to pulse the speaker. So it's still being used as a switching device. Mosfets do not increase power, or 'amplify', as you are thinking, they are still acting as switching devices. I wish I could explain it better, but it's not like I went to school for this stuff or anything.

..and stacking works, because there is now 2 mosfets for the current to flow through, so you've cut the resistance in half.

I come from the angle of playing guitar, and I build my own effects, and know how to fix amps. The more I study the RC cars, the more I realize how many similarities there are.

Here, I'll even explain further. See transistors and mosfets have a power source. If there is magnetic interference around a coil (Such as a guitar string, magnet in a microphone, or a radio crystal) it will produce a very small current, when that hit the transistor it allow the current from the power source through, which powers your speaker, or motor.

[slider69]

Sounds like ive been an idiot....Ah well... There has to be a way to amp that thing. I'll try to find out more tomorrow. And i did look at the board some... i got a partial number on the part... it was
15** and there is a symbol that could have been just a rectangle or might have had something more written on it... I scraped very carefully, too... u did really good with yours on that. I would be very careful on the board, as that involves shorting out with the LED's small PCB if u expose traces and solder pads...but i figure u knew that already....

[crazydave]

I've got it! Thanks man, you forced me to think about this, and I'm actually building off your idea.

Ok so you take a lead from the motors foward, you put some resistors on it so it keeps the current flowing to the motor, but(and here's the tricky part) you keep the resistance low enough that when you reach full throttle a little juice gets through creating a signal to switch on the mosfet, which then would turn on the juice straight from the battery bypassing the PCB leaving only a mosfet in the way.

In theory I don't see why it wont work, in reality I know there a lot I don't understand so it might not work.

I'm not gonna go killing my SE right away, but hopefully someone with better electronic knowledge than mine might see this and be able to build off of it. In the meantime I'll try to work it out in the back of my head.

[Nathan]

That's about how the zipzap NE (normal edition) dragster worked-it used a 2n22222-whatever transistor to pump current strait to the battery from 3 seperate cells. Of coursre, there's no reason why you couldn't use the stock battery instead. You need to know how much voltage the black box puts out, and how much current the mosfet can take to choose the appropriate resistor-the problem is the pulsing makes it hard to measure, unless you have an oscilliscope.


But, this idea would work, and has been done in other forms before (no reverse mod, 4.5v dragster...).

[slider69]

What does the black box look like to u? I dunno but i think it could be a coil of some kind like a relay or transformer....and i noticed something on the bottom of the board... they use 2 of the transistors stacked....did u notice the ZZ se seems to have more forward power than reverse (it did in my friends at least). I really need to check my old electronics book on the transistor...it does amplify current in some way..that's why transistor radios were made instead of using germanium diodes, right?

[crazydave]

The switch from tubes, to germanium transistors, and and later silicone transistors is mainly due to cost effeciency, and mass production. Many audiophiles, and guitarist still prefer tubes for sound quality, though silicone transistors do pass a current more effeciently..

Anyways, I have to look for those stack transistors, replace those with some stacked FETs, and we're in business. Good find! I'll test that after work today.

I have no clue what the black box is, I could only guess.


****EDIT: There are no stacked transistors on the bottom of the board the only single transistor in that vicinity is part of the circuit going to the antenna. My wife snagged the camera, so I can't take a picture of what I'm seeing. Is there anyway you can take a picture of your PCB so I can where those stacked transistors are at? Also do you have the Savannah, or Enfini?

Also I just noticed what you said. Diodes just keep a signal going one way, like a valve. They perforn an entirely different function than a amplifying/switching device. Transitors replaced tubes, and were much smaller, hence all the kids in the 50s and 0s with their little transistor radios. They were initially made from germanium, but later switched to silicone because it's more readily available and cheaper. Transistors do amplify, but not in the sense that you are thinking. I don't know how to explain it other than how i already did.

They amplify by opening up the current from your power source. mosfets just allow more of that current to flow than a silicone transistor.

[slider69]

Ya know the little black rectangles with three connections on the bottom (same side as the two circles of black epoxy that covers the two microchips...) They're the same size as the transistors on the MSpros but they have a different id number....There is one with the same number as the rest it's BKD i believe, (all on bottom are BKD or 3E), but it is raised up two times higher than the rest... It's the white RX7-- Savannah..... It has to be stacked, otherwise the part id number would be different vrai ou faux.....

[crazydave]

Every brand of car has different markings on the transistors. ZZs are different then MSs. I checked 3 different ZZ boards from each generation. In all the markings on the NPN transistors are 'S1A' with a sideways 25. There is a larger transistor on the back of the board labeled '3E' that is part of the reciever. On my SE the only transistor in the vicinity of the 3 contacts from the black box is also labeled '3E'.

If you follow it's leads on the circuit, you'll see it eventually leads to the antenna, and is part of the reciever.

Look at this picture of my PCB, and tell me if there is another transistor beside the one I described. The shaded area is where the black box is, the transistor is above that in the picture.

If yours does have an extra, you have got to get me a picture of that.

[baleisen]

Your discription of xistors isn't bad. Regular xistors vary their resistance in response to the voltage on their base. Like a water valve. They are there to control a larger current. Not cause it.
FETs are pretty much on/off. They have very little controlling ability. Like the 1st zip zaps, they are on or off.
BAE

[slider69]

I cant see if it has them in ur picture, in the savannah my friend has, they would be under the red and black wires in ur picture. I'lll post a pic when i find a digital camera around this place....

[crazydave]

Hey slider69, guess what? You killed my car you owe me 25 bucks!











Just kidding....YOU are da man!!! Found those stacked transistors, swapped them out for some FETs, and my car not only still worked, but has noticebly increased performance. Not as mind numbing as a fully FETed bit, but a SOLID increase in performance.

You hear that everybody....I'm sitting on the worlds fastest SE at the moment!:cool:

Thanks man, I think you made a great discovery.

Here's the ones to swap out. (highlighted in the picture)
http://s2k-kidz.com/crazycars/sefets.jpg

...and I'm more than positive the one next to it will boost reverse. That will be good to help conseverve battery life.

[vtheo]

yah!!!!!!!!!!!!!!!!!!!!!!! cool beans, have you tried with the motors that wouldnt work with it before? it seems like the alloys give a good prefomance boost too, but in handling... lets see a fetted, alloyed (hehe) 4.2 SE! :D

[crazydave]

Well I still have the ZZ 3.4 in it, but I'm running the green 9.86:1 gears with it. I'll tell you this much, when the battery died, and it wouldn't move in reverse, it still had power to move its self foward.

I did this kinda late last night, so I haven't had time to shake it down, but based off that alone, I'd say that it should pull a 4.2 easily. We'll see.

[slider69]

Great....now to get my friends.....Did you use NPN or PNP mosfets? IM guessing NPN...but i dunno... that's AWESOME, crazydave! u only replaced the stacked ones....there are others and ill prolly try them later....

[TypeZer0]

wut's the partnumber if the FET's u used? can u show a more detailed pic? thnx

[bdebde]

If it works better with 2 stacked fets, then stack up 3 or 4. That should increase power to the motor. Stack the reverse one for more juice there too.

[Bad Company]

Man, it's times like this when I wish I was more of a techno-electronics minded person. You guys are whiz's when it comes to the performance mods on these things. I guess I'm gonna have to get my feet wet and try this out. I'll just put $25 to the side for when I burn the PCB up and have to replace it.

[micro_Amps]

Good stuff guys,
Crazydave looks like you are on the right track here, the ZZ SE may have a pulsed single transistor that varies the current to the motor.
I am itchin to get my hands on one of these to see what the hell is on that pcb, should be only a few days away now.


:) uA

[slider69]

Im gonna try this on my friend's with a few changes....I want to replace all the ones (except 3e's) with mosfets and see what that does.....What do u think microAmps...

[micro_Amps]

Well, be careful just changing them without first knowing exactly what the circuits are for. If you change a transistor with an incorrect polarity FET it may just not work, or worse, may damage something. Also there are transistors there that are not used as switches but amps in the RF circuit, be careful with those, you cnat change those unless you really know what you are doing.
In any case, post your results here, and goodluck.
I am still waiting for mine to arrive, should be interesting when they get here.
:) uA

[slider69]

Those are the 3E's i believe in the receiver amp circuit.....The others appear to be wired in an h-bridge..i have to wait a while and double check everything before i go. but ill let you know how it works...btw...how r the airplanes going?

[micro_Amps]

This is a bit off topic, but the planes are going well. I have one that already flies very well, and one that will be test flown on Monday. Both are indoor flyers (basketball stadium) and both are BCG r/c controlled (fet modded) with KP-00 motors, which are about double the size of BCG motors. The planes are tiny, about 30cm wingspan, of foam and balsa construction with Li-Poly cells powering the motor and r/c. Amazing to watch.
I will be putting pics and info on my website, I just havent had the time recently. Check every now and then, I'll get them on soon.
:) uA

[malform]

I think we would all like to see a pic of the moded pcb crazydave.... And if Im not mistaken, this didn't effect at all your proportional throttle?

[crazydave]

The picture above is the moded board. nothing to see, the mosfets swap right it, and look no different than stock.

There are transitor in that area forming an H-bridge. It would seem the proportional signal is generated, and then the h-bridge switches it between forward and reverse.

The H-bridge is arranged so that it uses only NPN transistors, so I was easily able to swap them all out, but I wasn't able to get any gain in performance beyond the initial stacking of the 2 on one of the forward ones.

I believe this is because it's digital. Digital is a series of 0s and 1s, or off and ons. Hence the pulsed motor current. The yet unproven theory I'm working with now is that we need an A/D/A converter with a faster processor, to minimize that pulsing, in order to get more power to the motor. First I've got to figure out if that black box is an A/D/A converter.

I'm just kinda taking a stab in the dark, I'd like to hear your thought on this, Micro-Amps.

[bdebde]

crazydave,

Interesting theory on the da converter. Have you tried the batteries? My thoughts are that the smaller batteries put out less amps. Anyone who has done both the laymans button cell mod with 2 or 3 cells and a 2 or 3 cell with the old standard cells can tell ya that there is a big difference. A case in point, I used the batteries (stock dual) from a nakamichi racer (appear the same size as the SE bats). I put in a fet modded bit pcb and it was faster than a single cell with fets, but not so hot on torque. It certainly was not as fast as a dual cell with a stock pcb. I thought I had done a bad job with the fets or damaged something, so I took the pcb and put it in a single cell car, and presto, it was just like any other fet modded car. If you can try the old standard size cells in there, it may make a big difference. Even if this can be tried "on the bench" and use a meter to check amp output, that would answer that question. I have not tried this yet, as I only have one SE here at the moment, and am keeping it stock for now. I sent the other one to micro_Amps for his expert opinion. I actually sent 2 to him, one for me and one for him. If nothing else, he has bragging rights for the first SE down under, lol.

[crazydave]

That's a good idea about the batteries. Thing is are they 2 1.2v 70mah cells, or are they 2 2.4 35mah cells? That bench test would be a good idea, but my stupid cheapie multi-meter doesn't read past 200ma, and the batteries output more than that. You'll have to try that, and let us know what you find.

[urbanavenger]

*I have been following this forum for about a year but this is my first post. I tried the Layman's Cell mod on my ZZ SE. I went to a local battery specialist in here in Michigan and had them custom weld me a 4 button cell with 4x1.2 volt 40 mAh batteries for a total of 4.8 volts! The batteries were made by Varta and are the exact same batteries mentioned in Layman's post. I am sure of that since I also ordered the Layman cells.

I was really happy to get the custom battery when I saw it was welded and shrink wrapped and best of all It was almost the exact same size as the stock 2.4 volt battery. In fact when I fit it in my ZZSE there was no modification required.

So what happened... The proportional steering and drive both seemed to work. When I held the Zip in my hand and let the wheels spin it seemed to fly. However as soon as I set it on the floor and went I realized that the ZZ was SLOW. It seemed as if the ZZSE now had no torque. At first I thought that I just needed a charge. So I promptly went to a friends with a R/C hobby peak charger. I charged the batteries at .5 amp. When I attached a lightbulb and then a 5V LED directly to the battery the lamps lit up brightly. I definitely had good power. Yet still the ZZSE was slow (about half of the normal speed, with stuttering).

I believe this problem may be related to one of the components on the PCB operating out of spec. Over Volted. I'm guessing that the part that was faulting was the A/D/A converter itself. But that is just a guess.

Long story short I think we need to get these new ZZSEs faster. I loved the results achieved when adding 1 or 2 extra batteries to the original ZZ and I am really hoping that we can get the same or more speed out of the ZZSE.

[Jtskty]

Im thinking those button cells arent giving enough current out.

[urbanavenger]

I was thinking the same thing Jtskty... but when I attached the custom 4.8 Volt battery up to a test light it really lit brightly. When I attached the 2.4 Volt stock battery it lit also but much dimmer. I know common sense should tell us that more Volts should make more light. But it just seems that if the battery lit a lightbulb so brightly, the same battery should make the SE faster also.
Another consideration is that the stock 2.4 Volt battery is only rated at 80mAh. The custom battery is 40mAh. So if anything I was expecting half the run time. Thats why I tried to peak charge the custom battery. I would really like to understand why I experienced the stuttering and slowness. Maybe even the reciever was faulting out, Im not sure. Can anyone suggest an electronics guru that I can send the battery to for testing?

[bdebde]

Here is what I found doing a quick test on some batteries. I used a light bulb for the load.

Stock dual cells from the nakamichi racer (look same size as SE batteries) 80mah
Inrush upon connecting: just over 300 ma
steady load: 220 ma

2 150mah cells
Inrush: just over 400 ma
steady load: 270 ma

Looks like there is a difference, of course this was just a quick test. I will need to get things set up a bit better and do several of each type of battery to be conclusive.

[micro_Amps]

Guys, the general theory with nicads and NiMhs is that the larger the capacity, the larger the inrush upon connecting a discharged cell to a constant voltage charger. (delta peak charging current is different). So your results BDEBDE are correct and to be expected. The button cell pack at 40mAH is small and wont like being charged at 12C, 500mA (12 x 1 hr cap. charge) but so long as it is being charged by a good delta peaking charger there shouldnt be too many problems there. Keep in mind though, that the lower the charging current, the more power will be stored before the voltage goes delta negative (charger switches off). So if you've got the time for it set the charger at 1C and let it go, it will fill the cell almost completely. (Complete charge usually at around 1/10C) but takes ages. Keep in mind also that delta peaking can be less reliable in small cells at high charge rates. There is a higher risk the cell voltage wont delta at all and the cell will just fry.
At 4.8 volts there is a chance that the electronics in your ZZ are saturated. The car may run in your hand, whilst drawing almost no current, but as soon as you put it down and current is needed to be drawn in order to run, the controlls say 'no thanks' and you get the stuttering and lack of control. This is just a theory, but I have seen exactly the same thing on a modded new super BCG.
It sounds like your cells are ok, but the electronics doesnt like the extra voltage. I will be able to confirm this when my ZZ SE arrives (soon I hope)
Have you tried running it as a 3 cell?
I will post results along these lines when I get my car.
Cheers,
:) uA

[crazydave]

I was thinking with that four cell pack, maybe if you wired in parrallel 2 pairs of batteries wired in series, so that you are still putting out 2.4v, but double your mah. How fast the battery discharges also depends on the battery's internal restistance. Maybe having the the pairs wired in series will help cut the battery's restistance, because the current is being drained from 2 batteries at the same time.

Just another theory.


...but I personally, IMO, I don't believe the batteries are the problem. Figure the stock Tomy batteries are only 50mah. NiCds do have a lower internal resistance than NiMH, and the Tomy batteries seem to take more than what they're rated for, but still from that point of view 70mah for the Zip Zap battery doesn't seem that bad.

I'm gonna focus on the PCB for now, but I was thinking, we could always dual cell the SE with 150mah cells, and leave one hangin' out the back like we did with our bits, but I'm trying to avoid that, because I always prefer to keep everthing tucked away inside the car.

[bdebde]

micro_Amps,
I may have confused you with my last post, or perhaps I am confused by yours. I was not talking charge rates, I was talking dicharge rates when connecting the batteries to a light bulb for a load. While your theory holds true for charging batteries, I suspect a 50 mah tomy battery will discharge at the same rate or even more than the 150 mah ni-mh bats, for the simple fact that they are nicads (higher initial voltage). That would be a smaller capacity than the SE batteries. I may try to do some more testing on some different batteries this weekend. I think the physical size of the cell does come into play somewhat.

crazydave,
I definitely agree with keeping everything tucked inside the car!

micro_Amps,
I sure hope those SE's show up soon, it has been a week today. Then we can see what kind of micromagic you can work with these babies.

[micro_Amps]

Sorry bdebde, I did misread your post, I see now that you connected them to a light bulb. I guess I focussed on your use of "inrush" and I took it as a charging term, then I went on rambling about charging theory. I usually do my posts at about midnight, I guess my eyes (and brain) were starting to close.
You are right, the nicads will flow more current than NiMHs on the same circuit simply because they have a lower internal resistance allowing more current to be pushed out.

I am expecting the cars on Monday, I find it usually takes a week plus a weekend, then they turn up. I will keep you informed.
:) uA

[crazydave]

Yeah, I also hope you get those cars soon, micro_Amps. I can't wait to hear your point of view on these. :)

[slider69]

I just did it....after 3 good hours with my best friend (A 15 watt soldering iron) I completely mosfetted my Enfini. (I just got it yesterday...had it for about three minutes before i mosfetted it) I put 2502s (i think there are 5 counting the 2 in the stack) on the side without the black coating, and replaced all the transistors coated with epoxy with 6401s (there are 4 on that side)....It doesn't go as fast as a regular zz tho....think there's something else holding it back...i am using a 3.8 with the blue ratio....so it should accelerate a lot faster......what do you guys think?

[micro_Amps]

If it doesnt even go as fast as a regular ZZ, then I wouldnt call it a success.
It is supposed to be an improvement, not just a change for the sake of it.
Do you know the actual purpose of the components in the circuits, why are there 5 fets?, I would have thought it would be an even number of fets.

:confused: uA

[slider69]

They are in an H-bridge i believe..there are four places for a transistor, but two are stacked......I think it's my battery tho...the se doesnt want to charge now....I charged it up and it ran for ten seconds b4 the battery went down.....but that could be the ADA converter wasting energy or something.

[slider69]

Crap man...i take everything back...I messed with my connections a little bit...did nothing....but i changed the batteries on the controller......geez...this thing is a lot faster than a stock se....but it still lacks the torque and acceleration of my zz with msizer pcb......thats gotta be the digital to analog converter you spoke of....oh...btw.... i think the converter is under the new blob of epoxy...(the original zz only had one for the receiver chip)....

[crazydave]

I feel like no one read what I posted before.

There is an H-bridge on the output. (It's circled in red on the pic).

http://s2k-kidz.com/crazycars/sefets2.jpg

First I replaced the 2 stacked transistors with a FET, that improved it. Then I stacked that one, and that further improved it. Then I fully FETed all four of those, and it improved reverse, but forward only seemed slightly improved. So I stacked the one right next to the forward one that was stacked, and go reverse up to par with forward. This was testing with a Perfection 4.2, and 9.86:1 gearing. I at least got it pulling the 4.2. Acceleration is not up to par with single speed ZZ, but it is greatly improved, along with run time, making it a somewhat enjoyable car now. I think after you get yours, micro_Amps, you'll qualify that as at least a partial success.

Now, slider69 is right, there seems to be something else holding it back, and it is my theory that because these are advertised as 'digital' proportional there must be an A/D/A converter in there, (I'm guessing the big black box) and since these are just 25 dollar fully proportional RCs, I'm guessing it's a pretty crappy A/D/A converter.

Now I'm looking for evidence to support my theory, since I know very little about digital other than it pulses the current with all it's 0s and 1s clicking on and off.

[slider69]

i dunno. the big black box looks like a coil i saw in a coffee maker's timer circuit once....the thing i notice different than a regular zz is the second blob of epoxy that would cover up a small dip package.......this is usually the format for most counters, operational amplifiers, etc. , right? Or am i totally off? There is also a circuit on the back(under black paint) that has something to do with reverse....I know because i staticed a PNP fet out and it lost reverse until i replaced the fet in question...mebbe if you stack another pnp on the one that corresponded to forward.........

[bdebde]

I'm still thinkin the battery is holdin it back!

[slider69]

Dunno...definitely does on run time i figure....But something else is definitely screwy....it's a friggin 2 cell and it doesn't run like one...the weight is no excuse for a lag this great.....Gotta be the analog-digital thing...the question is how do we convert?

[bdebde]

I'm telling ya, those smaller batteries don't put out as many amps as the larger (physical size) batteries. If we could get some of the older style batteries in there, I bet it would get movin a lot better.

[micro_Amps]

Got the cars today BDEBDE, thanks very much.
Stay tuned................
:) uA

[bdebde]

Cool, hope to hear some good news soon!