LM393 acting like an Amp - Hysteresis

[kinarfi]

Any one familiar with my posts knows I've been playing with this power steering unit for quite while, I'm still trying to smooth off all the rough edges.
**broken link removed**
2014-12-25 POWER STEERING - Hysteresis.asc is the whole schematic with Hysteresis which will not sim.
2014-12-25 POWER STEERING - Hysteresisplay1.asc is a stripped down schematic with Hysteresis that will sim.
2014-12-25 POWER STEERING - Hysteresisplay2.asc is the same stripped down schematic with out Hysteresis the will sim.
The actual unit has a problem with one of the Gate feeds to the Coil driving FETs (Now changed to transistors), the feed on one side does not switch cleanly from on to off or off to on but hangs in the middle and oscillates, it did the before I added the Hysteresis also, but also had lots of spikes from the PWM. I also noticed that the hysteresis shifts the center of things to a higher voltage, as shown inb the .png attachment.
Does any one have any suggestions on how to make the Gate drivers switch cleanly?

 

[alec_t]

Does the actual unit have gate stopper resistors or ferrite beads (e.g. 100Ω/bead immediately adjacent the gate/base pin)?

 

[kinarfi]

I use a 10Ω resistor, but I have about 3" of wire to feed the gate via a connector pin, notes on components in photos refer the schematic 2014-12-25 POWER STEERING - Hysteresis.
Hope this makes sense.

 

[MikeMl]

The .asc files are missing lots of library components. You will have to zip up all the subcircuit files you are using if you expect anybody to work with the .asc files.

 

[kinarfi]

My sub files, plus I changed some components

 

[alec_t]

I use a 10Ω resistor, but I have about 3" of wire to feed the gate

Try increasing the resistor value and/or reducing the 3" to 3mm.

Edit: BTW, it's a pity so many of the component designators are missing: it makes it difficult discussing the sim operation.

Edit2: When you say the whole circuit won't sim (I haven't tried; it's like a jigsaw puzzle assembling the missing bits), which part gives the problem? The comparator section per se with hysteresis seems ok.

 

[kinarfi]

Can you send me a "print screen" of what comes up for you, so I know what's not coming up?
Thanks, Kinarfi

 

[kinarfi]

I just installed a new LTSpice and opened a version of my powersteering schematic that I had been trying to get to sim and it couldn't find ina122new, rly_dpdt, LM339, potentiometer, or reg3all. I can certainly see how that could be a problem.
What would be a good site to reload valid sub circuits and assemblies that others would most likely have. Places I know are ltwiki and Yahoo groups, LTspice.
One thing I've wondered about, is there a cross reference to all the devices that come in the LTspice libraries? for instance, LTwiki shows how to modify LT1017 to be the LM393, nice to know. Any way, I'm working to clean up my files, it will be a long term project.

 

[alec_t]

Can you send me a "print screen" of what comes up for you

Yup. Here it is.


Edit: Apart from the missing components, your LM317 symbol is a non-standard size?

 

[kinarfi]

Thank you, I will redo the .asc and post it, please be patient.
I have a bad habit of editing components to make them fit my wants, needs and desired, like adding that little circle on the LM317 and making it smaller, I guess I'll quit changing the size and just add a J to anything I modify.

 

[kinarfi]

http://ltwiki.org/index.php5?title=Components_Library_and_Circuits
scroll down just a little and find
A Large LTspice Folder from Bordodynov
This has files I use like pot&sw folder for pots and it has Pote.lib for the .sub files

 

[kinarfi]

In my schematic for my power steering design, I have a LM7810, but it seems that LTspice doesn't like some of my models so I'm trying to use what LTspice supplied when newly installed. There are no 10 volt fixed regulators, that I could find, in LTspice, but some where I saw a " ROLL YOUR OWN " page, then i couldn't find it again, so I thought I'd give it a try anyway. Here's what I came up with.
My schematic still won't sim though, it works ok with just the INA122, gets real slow, when I add the first NE555, and fails to understand the INA122 when I add any more.

 

[alec_t]

I find it helps not to sim standard functions, and to sim only a section at a time, if you want the sim to run quicker (or even to run at all). For example, I wouldn't bother simming a 10V regulator; I'd just use a 10V voltage source. Do you really need to sim the 1NA122? Won't the standard ideal opamp do?

 

[kinarfi]

The INA122 needs to be in there because it has the 5 V reference for when the 2 inputs are equal and with a gain of "10", a -.1 volt diff at the input creates -1 volt (4 V) at the out and with +.1 volt diff at the input, you get a +1 (6 V) at the out.
The good news is, with the reloaded LTspice and the use of their devices, subs, and .Libs, it will now sim, except for the relays, I think they have too slow of a response time. You most likely won't have the LT1086-10, I had to build my own by manipulating the LT1086-12 files as explained earlier. One thing I did encounter was that if I use Excel to do my modifications, they failed to work for ever after. If you do decide modify a .sub file, I found the best way do it is to find it with Window Explorer, right click it, and open with the spice program. There you can copy, paste and change it into LT1086-10. FYI, LT1086-12.sub is embedded in LT1083.LIB.
To make the LT1086-10.asy file, open the LT1086-12.asy in C:\yourSpice\lib\sym\PowerProducts with word pad and make your changes as shown above and save as LT1086-10.asy

Have fun!!!

 

[alec_t]

The INA122 needs to be in there because it has the 5 V reference

You can easily sim a 5V reference with the standard voltage source.