Another PCB mod and Schematic

Dayfall · #1 10-21-2002, 12:38 PM

- This is my first ZZ mod and my first time thinking seriously about learning SMD.

- I wanted more power for my ZZ, and I have seen lots of good mods. Since I want to keep reverse, I decided to monkey with the H-bridge. I decided to take a look at the circuit board and try out this mod, http://tinyrc.com/forums/showthread....threadid=1736. The resistors to be bypassed were not connected to what I expected. They seem to be going to the gate of the mosfets. After i checked the voltage levels, I suspect that the transistors are not mosfets but are NPN and PNP transistors instead. There is a voltage doubler on the board (which can be used for other mods!) but I doubt this is for operation of mosfets.

- I couldn't find a schematic of the H-bridge so I have included my best guess as to the SMDs. R20 and R21 go to the low side transistor's bases and R16 and R19 go to the high side transistor's bases. R17 and R18 are pullup resistors (For stability?).

- Now for the mod: I highly suggest that the four driving resistors be lowered. I have gone ahead and shorted out R16,R19,R20 adn R21. Originally I could pull 80mA through the circuit. With the mod, I can now pull 150mA! The best part is I am only pulling 166mA from the battery. So I get over 4 minute run times and my ZZ can now run on carpet with red-red. Yaaaa! I also tried to give more power to the steering coils but their resistance is so high that I could only pull an extra 13mA. So I might change the magnet to a higher quality one instead.

- Here is another thought to those that really wan't a challenge. Use N channel and P channel mosfets that opperate at 3 volts; install another battery for the voltage to run it and you will then have a propper H-bridge.

- Anyway, my car runs better than ever. If anybody else experiments with changing the resistors or adding more transistors (you still can't get around the 0.2 voltage drop, I dont think) then please let let us know the results.

marc515192 · #2 10-21-2002, 04:43 PM

I'll try this mod on my other zip and see what happens.

Thanks for the tip.

Viper76Man · #3 10-22-2002, 12:11 AM

If your schematic is correct (and I suspect it is) that would explain why people are seeing mixed results by only shorting out the 68's in the H-bridge and why you are seeing much better by taking care of the 220's also.... nice work...

Viper

Dayfall · #4 10-22-2002, 05:23 PM

- I have some low voltage SMD mosfets that I thought should be tried on my one and only Zip Zap. Naturally I am extra careful not to destroy it. So, after verifying voltages and currents I disconnected a NPN transistor and replaced it with my mosfet. And I could not believe it. It worked! So I tried two mosfets and got even better results.

- Measuring from battery negative to the blue motor wire, running the motor forwards and holding the wheels this is what I got on the oscilloscope:

Original configuration: 0.42v
Shorting R20 and R21: 0.35v
Replaced NPNs with N-Mosfets: 0.2v
Stacked 2 Mosfets: 0.09v
(I suspect that 3 Mosfets would give even better results, but I have gained 0.3v and that is enough for me)

The advantages to this mod are more voltage to the motor, less power wasted in the transistor, and less strain on the IC. My mosfets were pin-for-pin compatable and only cost 44 cents each. The values of R20 and R21 have no effect on the performance. It should also be noted that apparently the IC can sink and source current, so there is no need for a pulldown resistor to turn the mosfets off.

- The mosfets I used were N-Channel 25v SOT-23 from Digikey. But I somehow don't expect too many people to try this mod out.

Oh I almost forgot, I can pull 260mA from the H-bridge. But I have no idea why anyone would wan't to do that.

john · #5 10-22-2002, 09:45 PM

I am so interested in this procedure, I must say. I understand how to short the four resistors, and I know which they are, but I totally do not understand anything else you are saying. Is there a way you can post that info in layman's terms? I'd love to try it, but I'm electrically clueless.

Nightwing · #6 10-23-2002, 01:08 AM

Try this: It may not work or cause problems.

Using above diagram.

Disconnect R18 & R17 ( Remove )
Leave R19 & R16 ( or cut in half or double value )

And either short or cut ohm's for R20 & 21

The problem is you dont want too much base current or will fry the transistors.

Now is the + voltage comming strait from the battery or through another resistor?

Both R18 & R17 are biasing the transistor and not letting it fully switch.

Tnks Dayfall for the part number.

PS: Dayfall. Have you try a cobo of P-Channel on top and N-Channel on the bottom?

Dayfall · #7 10-23-2002, 10:28 AM

It does seem like removing R18 and R17 would help. I havn't noticed much loss across the upper transistors so I havn't bothered with them yet. Now that I have more current going through the motor, I might should go back and look at how much current is going into the transistors gate. It is probably to much since 10mA would be way more than we need.

Well, I just checked it and sure enough, 50mA. Woops. I am sure it can take 100mA. but that reading seems kinda high. With the 68ohm resistor it takes only 5mA. I am guessing 10mA is what I need. 30ohms is probably close to optimal. The problem now is to find a resistor small enough. Regular SMD resistors are huge compared to the ones in the ZZ.

The reason I don't think P-Mosfets will work is because the maximum voltage avaliable to the gate is 1.2v (for those of us with only 1 battery.) Yes, + on my schematic is battey positive. I just looked at the charts for my mosfets and it says if Vgs=2.7v then Rds=0.24ohm. I am betting that the resistance will be way too high for 1.2v.

John: I wouldn't suggest trying this mod unless you have a good soldering iron and steady hands. The procedure is easy though. Just take off the two transistors and replace them with the new ones. Here is a picture of what I have done so far:

john · #8 10-23-2002, 10:37 AM

Hmmm. Okay, I think I'll wait a bit on this one. I can solder, and I use tools to steady my hands (a surgeon, I am not

) I am just clueless about what little electical part does what. I'm learning though. I've been scouring the web for info on DC electricity. I do the electrician's work in my house, but it's much simpler to wire in an AC light fixture or an outlet than it is to hotwire a ZipZap.

I think I'll start with something a little easier... like headlights that turn on when the car is going forward or somthing. I just read how to do that, so perhaps I'll start small, and just watch this for a while and see what happens.

Thanks for the subtle warning and the pic, I'll heed the warning, and study pic!

Nightwing · #9 10-23-2002, 11:46 AM

The only reason I suggested the P-device is the top of H is the "source" and the bottom is "sink". So you use P's on top an N's on the bottom. Im thinking of max volts/amps right now.

Wait a minute... Wonder if they used NPN for both! Usualy the setup is PNP on top and NPN on bottom!

Check around the PCB. They may be running a "switching inverter" PS for the receiver and decode IC. The chip set need at least 2.2 volts I think to run. If so there may be 3v running around that could be used.

PS: Good job!

Dayfall · #10 10-23-2002, 12:23 PM

Yep, there is a voltage doubler on the board. It is labled A3_3V. The output seems to be the center pin between the other two. Right now I am trying to steal from it and run some lights, but I don't think I can pull more than around 50mA or so without the noise interfering with the RF stuff and IC.

The reason the N channel works is because there is 3 volts coming from the IC. The Source is connected to ground and so Vgs=3v. We normally want more but you know... The upper side of the H-bridge is diffrent. Assuming no voltage less than ground, we would need 6volts to the gate, that is because the motor would take 3volts away from the Source. That is if we used and N channel mosfet. A P channel one would need either 3volts on the Source (meaning running the motor at 3v) or putting -1.5v on the gate. Either could be done if someone wanted to.

I think the upper transistors are PNP because it looks like the intention of the circuit was when one side of the motor went to ground, that would pull down the gate of the transistor on the other side. And since that is the other transistor we want to be turned on I think it is a PNP type.

Also there is somthing funny about the current going to the PNP transistors. Even with 10ohms instead of the 68ohms I still don't get the power I do when I short the resistor out. I am going to try a PNP darlington and see what happens.

lowerdfool · #11 10-23-2002, 01:34 PM

You guys seem to know a crap load about electronics. I'm guessing you went to school for this but do you happen to know of any books or anything online that explains some of this stuff. I was going to go to school for this but I may be losing my job that would pay for the schooling.

Nightwing · #12 10-23-2002, 02:17 PM

The base design seems like most designed on the TX2/RX2 chip set. They usualy will put a micro switching PS to stablize the voltage feeding the regenative receiver and the RX2 chipset on the board, since voltages range goes all over the place. I agree would be hard press to pull much out of that circuit. Could beef it up.

Wonder if switching from the 68ohm resistors cross to a more TTL scheme of flipping would help?

The thing is what is the exact different between a Bit-Chrg-G H bridge and a ZZ one? Got a MicroQ and my Tomy Bus is still on order so could check with the Micro-Q varrent.

PS: lowerdfool. Did in high school but work with this over a lot of years. But mostly reading, doing and working at companies. One of my books I always keep on the shelf a ARRL handbook. They have a new one every year. Has base info. Check around grab one used. The new one are expensive but the theory has not changes. Proabbly no further back than 1974 or 75. Else you start into tube therory.

Check Radioshack they have some help books if they still stock them.

PSS: Got to run out to a client to demo a pice of SW I cook up for them. They was a site I ran across a few months back on using a PIC Processor to add multy speed to a bit but cant remember the site right now. From Japanes I think. It has the diagram of a Bit H bridge I think.

tinyrc · #13 10-23-2002, 09:40 PM

What FET are you using?

Gruder · #14 10-23-2002, 10:48 PM

DayFall: Good Job.

Just wanted to let everyone know that they can get the parts online at http://www.digikey.com/. Also I was woundering what % of increase you seen after doing this mod?...

Gruder

Nightwing · #15 10-24-2002, 12:37 AM

Finaly got back in. Wow love putting out fires and bored clients!

The site I was thing of that did have the H configuration of a Bit Char-G H bridge and was a Japnaes site! Also it the guy that has the Japanes semi with the umpteen lights on it.

This is it under dual and tripple speed section on his site.
http://www.age.jp/%7Eex2/bit.html

Its basically the same but missing R18 & R17, direct connection to RX2 and the two cross resistor. A lot cleaner.

Could it be the spects on the transistors are different? Be fun to take a "set" from a bit pc board and replace the ones on a ZZ.

Since I am taking the next couple of days off, will get to "have fun" also!

PS: Wondering how close to ground the NPN are really getting down to?