acondit: Are you planning on running this from emc? Pluto maybe? mesa? I don't remember what you where planning.
I think the way the uP on the uhu worked is it actually ouput pwm/not-pwm. The pluto or even the software pwmgen doesn't ouput an anti signal on the opposite pin while the other is generating pwm. (hope that made sense).
So if you look at your circuit if the right side of the bridge goes into over-current (trips the flip flop) it will not get reset until there is a left side pwm rising edge. This is why I or-ed the two pwm signals into the flipslop
http://www.electronicsam.com/images/...schemfinal.PNG
^ the logic was breadboarded and 'seems' to work.. But I have not found the time to make a board yet.
I have the pluto and it seems to work just fine. It doesn't really matter what size servos you use.. It really just outputs pwm and counts encoders. The pluto could care less what it is driving. The PID, feed forward tuning is done in emc.
If you haven't bought it yet.. mesa has a printer port module now - 7I43 for about the same price.. (much larger fpga) They have just added the hostmot2 driver to emc2 which gives a little more flexabillity. The driver for emc is pretty new and you would have to do some research. (like I don't think it does pwmup/pwmdown yet just pwm/dir). But like I say - the pluto has been working fine for me. (there have been some issue with a few computers that will just not communicate with it. (I have not run into this yet))
OK, I have attempted to make all the changes suggested by Kreutz and Sam. I have split the Vcc into Vcc 5v and Vdd 15v. I added a 4071 to OR the up and down PWM signals going into CLK on the 4013 flip-flop.
(Sam's schematic)
Sam inverted the enable signal going to D on the flip-flop and then inverted it again going to another OR gate for the Enable and Current Limit signals going to the reset R on the flip-flop. Additionally, It looks like Sam could have picked up the feed to pin 5 of the OR gate IC5B from pin 7 of the inverter IC3C and eliminated an additional OR gate. (Or did he need the additional gate delay?) (My schematic) I believe that the pull-down resistor R14 and Diodes D7 and D8 are supposed to take care of this. However, there are still unused inverter gates and unused OR gates that could be used to eliminate those Diodes (per Sam's design) if that seems to be a better idea.
I added a 7805 for the 5v supply that is fed off of the 15v output of the 7815. I don't think that it will even need a heatsink but I didn't actually calculate the current draw on it.
Alan
Those opto's are similar to the ones I used. They invert. There is really no way around it. The pull ups are required. These are considered open collector. The pull down resister is not going to do anything as there is nothing to pull it up.
sam
Hi all,
Pls help me on this:
I’m DIY a 5 phase pentagon driver base on schematic at post#107 in this thread
I can not find out some part in my country so I replaced by others:
+ MOSFET driver: IR2184
+ MOSFET: IRF540
+ Diode bootstrap: MUR110.
+ Gate diode: 1N4148.
+ Cap bootstrap: Polar cap 4.7uF.
+ PWM : 25Khz
After finished driver board, i run driver at V-motor = 24VDC no problem and driver work well. I continued to increase V-motor = 53VDC, some MOSFET was burn and Rsense (0.22 Ohms) die immediately
Pls advice any problem so that i can overcome this issue.
I will upload my schematic when I come back home and it is same schematic at post 107
Thank in advance.
Giang
Hi all,
here is my schematic, pls help me how to modify schematic so that i can run with high voltage.
Thanks
Hi all,
PCB top and bottom for checking
Thanks
The circuit looks OK except :
1. Add a .1u 16v cap to the power pin of each ir2184
2. Add 33 ohm in series with each bootstrap diode.
3. Add 10K between gate and source of each fet
4. A 500-1000u cap on the 12v might also be needed.
Hi H500
Thank for your help
I also receive a Kreutz’s comment in PM.
So I will update my schematic as below:
1. Add a .1u 16v cap to the power pin of each ir2184.
2. Add 33 ohm in series with each bootstrap diode.
3. Add 10K between gate and source of each Fet.
4. A 500-1000u cap on the 12v might also be needed.
5. Replace the bootstrap capacitors for 2.2 uF ceramic.
6. Reduce also the gate resistor value to a mere 47.
7. Eliminate the gate discharge path (resistor in series with 1n4148 diode).
8. Move chopper line from IN pin to SD pin on IR2184
"Kreutz wrote: Please, try replacing the bootstrap capacitors for 2.2 uF ceramic, normal electrolytic's will not work reliably due to high esr. Reduce also the gate resistor value to a mere 47 ohms and eliminate the gate discharge path (resistor in series with 1n4148 diode). Actually the pcb design is very important for a working drive and I haven't seen your pcb design yet. Please look at International Rectifier's application notes"
Hi all,
i rebuilt PCB follow new SCH, anyone help to advise
Thanks.
Hi all, yes I am digging up a VERY old thread, I happened to stumble back onto this looking for something else, but re-read through everything and am happy to say we had some big hitters comment here and yes I may be feeling a little nostalgic after a couple of quiet whisky's, but my hope is that some of the "new guys" may benefit from the information here.
Just to check in after ten ish years the H-bridge I mentioned earlier lasted for around 9 years on my first router, and now I have a second router with some better mechanical components that has been going for around 6 years, I have plans to retrofit a 2.5ton milling machine in the next year or so, time/disposable cash pending.
I appreciate all the help from my peers here on CNCZone, I always felt that I "kludged" my way through designing electronics as it isn't my day job, but if it wasn't for you guys my kludging wouldn't have worked at all.
Feel free to drop a comment about anything here if you feel like it, though we may never meet, I consider you my friends that helped me do better than I would have. I am sure the moderators will cut us some slack if the comments are a little off topic.
Cheers.
Russell.