K8055 - Dual Wiper Motor Control with Variable Speed

Thanks Tim: I just have to learn some more understanding diagrams, but I will get there! It would be useful to know more about your parts. Any npn-transistor? Any diode? Ratings on relay?

Tim: Yes, I know that you are using relays, I have been following the thread. Your solution is excellent. The above diagram is what I linked to in a previous thread. I do like your relay solution but I wanted simpler (for me) setup and the 130.00 is not really a bad price to pay for motor drivers for me. I will of course only use one until I have everything figured out. When I mentioned waiting for you to research using three drivers I meant using two K8055s to push three motor controllers in X-Sim2.

What I really enjoyed in this thread is the mechanical aspect that you have been discussing. I am playing around with the idea of a single gas strut mounted between the two sides of your lever system to support the weight of the platform, something like what Frakk has posted using leggos (also a huge fan of leggos but my daughter is now a teenager so I have no excuses for buying her more, lol). Instead of a spring on each leg I will use one strut mounted on crossbraces between the legs of each actuator.

Hi Historiker - I'm with you there... mechanisms are very much more in line with my background than electronics!
(I was definitely a child of the Lego generation!!)

I think Frakk has hit a great idea with the different position of the spring, and your suggestion earlier of a hatchback door lifting gas spring could work for this application - the only problem may be that those struts sometimes have a damper in to slow the movement which might strain the motor and limit the responsiveness of the actuator... be aware of the full range of movement that this actuator can provide!

If configured as Frakk suggests with his lego, the diagram I posted of my spring mount (the awful hand drawn picture) would work - as soon as I manage to get hold of a suitable spring, I will rig it up and post the results! (unless you beat me to it! )

I'm glad to hear that you are finding the post interesting!

Tim :

Ha! I doubt I will be getting anything finished until summer, my wife has a honey-do list as long as my arm. But I am enjoying designing and prototyping at this stage.

Very good point about the dampening feature of struts, I had not thought of that. I still like the idea of mounting return springs between the actuator legs instead of on each leg though...I will need to think on this more.

Again, keep it up, many of us are watching and hoping for your success! :

Easy enough Fred - there are only 6 components in the relay board, and 2 of them are the same thing!

I used the following for each relay circuit:

1 x DPDT relay (Maplin code N08AW) 12V, 10Amp if i remember correctly
1 x Relay Socket (Maplin code N32AW)
1 x 1K resistor
2 x power rectifier diodes
1 x NPN Transistor

I later added heat sinks to the transistors, and mounted a fan over them... other than that, thats it!

On the diodes and transistors, I am reluctant to advise my components as the ones to use, because they were selected purely on the basis of what I had available, and I have not yet got round to reselecting these components!
:

Have a look in the stocklist of your supplier, and select something more suitable based on what is available... There are so many components in the world, and not all will be available to you - see what you can get your hands on first.
By all means post the specs of components that you are looking at and I'm sure someone more knowledgable than me would be able to say whether or not they would do the job... I'll have a look too... I'm not an electronics guru, but there is one around these parts!

On that point actually, Frakk has offered an alternative circuit quite early on in the thread using Mosfets... I havent yet been able to get hold of suitable components - maybe you could try the upgraded version he suggests... It would be interested to see how it performs!

Here is his alternative circuit:


Tim

Sounds familiar!

Now I will order parts to make this driver, first checking if I can use some parts I alredy have. At least the waiting-time for parts can be used to shorten that list (exept it has a tendency growing as fast a its been shortened :uups: )

If I make this work with amc it will be a cheap solution! Might try also with mosfet, anyone think the mosfets from the dSMHB will work (irlz44 or irlz34)? or even the transistors? (bc546) They are NPN.The dSMHB have general purpose diodes, you are using power diodes, so I guess I need to order them.

Is 10amp enough on the relays? On my anmeter the vipermotors spike shortly up to 30 amp once they are started (stable around 5)

Is there a lower risk burning mosfet (or transistor) in this driver than the h-bridge? in that case, is it because switching direction is a critical part ? (Alredy have a fan ready from my dSMHB)

Fred

Hi Fred

The 10 Amp rating is for a continuous value with acceptable spikes - the wipers themselves are recommended for a 10Amp fuse, so I figured it would be ok for now!

When I eventually connect a car battery rather than the PC power supply I will be including 10Amp spade fuses in the build to protect the circuits

Tim

Small question in my lunch-brake: Is Solid state relays also an alternative in this driver? (more exspensiv, longer life and silent) Like this one http://www.futurlec.com/Relays/SSRDC100V40A.shtml well...they might switch only one direction!? :

You can use solid state relays, but you have to make sure the electrical operation is the same as a DPDT (2circuits/2position)switch to route current properly in 2 directions. You can alternatively use 2x SPDT relays.

The one you posted is equivalent to a SPST (1circuit/on-off) switch. You can parallel 2x SPST with one's input inverted to get a SPDT.
If one goes that route, it makes more sense to get 4 MOSFETs instead of 4 relays. But then it would make more sense to just go with a proper H-Bridge!

So here is the part-list I think about ordering to try this.

2x 2N2102 - NPN General Purpose Amplifier
http://www.futurlec.com/Transistors/2N2102.shtml

2x GR12V10ABL - DPDT 12VDC 10A Blade Terminals Relay
http://www.futurlec.com/Relays/GR12V10ABL.shtml

2x GRBLADE - Socket for Blade Terminals DPDT (10A)
http://www.futurlec.com/Relays/GRBLADE.shtml

4x BYW29 - BYW29 16A Fast Recovery Rectifier Diode
http://www.futurlec.com/Diodes/BYW29pr.shtml

A protoboard

I have resistors, and if the mosfet from dSMHB is suitable I might try with them also. And Frakk... I see there is mosfets in the solid state relays but i was not planning to develop a new mosfet h-bridge (bad idea :lol: )

Happy for confirmation on my part list. This will cost me less than 20 usd incl shipping, so its worth the try!

Regards Fred

Those transistors are rated for 1A, I hope you realize. If you don't want to use MOSFETs, you should get a 10A NPN.
IMO, it is easier to go with the FETs since you already have them, and worst comes to worse, you can order the regular BJTs too.

MOSFETs are not that evil, trust me.

The blade socket has screw terminals as I see, it will be easy to mount too. I would go with the DIP relays to save, but thats because I have no problem soldering.

Thanks again Adam, there was a reason i posted that list! :yes:

So the IRLZ44N mosfets I alredy have for my sDMHB is suitable to use in the diagram you posted then!? I am not afraid of soldering, after all I managed to solder amc with sucess, and I only use a ordinary solder rod. (not solderstation with temp. controll) I did not find dip relays with 10amp at futurlec, but will look elswhere also. If transistor: http://www.futurlec.com/Transistors/2N5881.shtml getting closer?

No plan to hijack this topic Tim, but I guess it is usefull also posting about parts needed Testing this driver with amc could have its own topic maybe...

Fred

Tim,
Any updates on your actuator armature and a helper spring?

Yes, the IRLZ44N is going to work good. It is logic level, so the 5V signal from the AMC should operate it nicely.

The 2N5881 is good, just be careful cooling the case, it is connected to the collector and will have some voltage on it.

I would get this: http://www.futurlec.com/Relays/JQX-115F-12.shtml
Similar in characteristics to the blade type, 8A isn't that much different than 10A. The short spikes will go well above 10A anyways, the average current will be much lower than that. You can connect them in parallel too, but that might need the direction signal to be amplified with a small NPN transistor.

Also, if you haven't already found some perf boards:
http://cgi.ebay.ca/Solderable-Perf-Boar ... 1786wt_911

Nice!!! I guess I am ready to order. To bad it takes 3 weeks to get it here. Hope this posts is usefull for others too. if this driver works well it will be a good alternative for rookies in electronics, and i hope less sensitive and easier to build than Thanos H-bridge. Maybe pololu drivers (my last option) will be better for some but it is fun to be able to make it 100% diy

Fred

- No problem Fred

Anything that improves the relay circuit can only be a good thing!

I am also really interested to see how the AMC compares to K8055 performance with the same speed controller circuit!

Tim

I'm afraid not yet Historiker... I am working in England this week and typing this on a train

Next weekend however, is a different story

Tim

I really envy the rail system in Europe, we are too dependant on cars here. When I lived in Germany I could hop a train to anywhere for the weekend and enjoy a book on the way.

Re: Connecting Position Indicating Pots to K8055...

It's not too bad I suppose Historiker - at least I get another hour each way to play with designs and read the forum!

Some advice for anyone using the inputs on a K8055 as per the connection diagram below:



I have found 10K linear pots to be much more reliable and less jittery than 100K linear pots.

Here is a video of my latest test, showing the PID control available by using a 10K pot connected as the diagram above:

http://www.youtube.com/watch?v=if0nzgkNcgk

A question to anyone that uses Thanos setup - how does the performance here compare? (I have no other setup to compare it against... yet)

Tim

Seem to me that there is a small delay but you also said you do not have full effect on the motors yet. It will take some weeks before I have parts to build this driver and test it with amc, and it will be interesting to compare, both different controllers and different versions of this driver (transistor and mosfet) And if there is a delay, is it in controller or driver?

btw. In your list you wrote power rectifier diodes, and Frakks diagram says schottky diode. I ordered both, and they are all rated for more than 12v.

Fred