Ward's Build of Bob Boyce's TPU

[HumblePie]

@All,

I borrowed a 'Modular Circiuts' brand PLD Programmer and programmed 4 GAL's today.  EDIT 28-Nov Unfortunately, the old PLD Programmer did not support newer Test Vector syntax I forgot it is the .jdc file that contains the optional test vectors, so I could not get auto test done.  The .jed file only contains fuse map information.   
BTW - A PAL or GAL works by cross linking logic gates very which way.  You switch off connections you do not want to leave only the desired logic behind... with or without clocks.  You do this by defining when outputs are true using Boolean expressions (you do not need Boolean Algebra, it is built into sofwtware for you to minimize equations).   

Anyway... Then I turned down borrowing same person's 'Logic Port 32' USB Logic Analyzer.  I was a fool.  Installing these wires took longer than installing that software would have.   Fortunately, my DDS 20 does 0.3Hz min (stalls at .2Hz).  I was able to verify all works exactly as the simulations in reply #1 show. 

Edit - /oe is somehow active high as in just 'oe'.  Making it low engages high impedence.  Yes it is a flaw, but it is this way for a while.  The opposite polarity caused constant tri-state and I forgot to change the signal name to oe.  It s an asynchronous input at least.

Go ahead and burn that 'JED' file posted there.  It is good to go! 

Next I will assemble 6 RC delay ckts between this GAL and a buffer / inverter chip.  I choose to just change the buffer to inverter as needed when I may run out of non-inverting drivers and need to use inverting.  Both chips socketed so no problem.  I could make equations to control this, but there is a limit to how many lines of equations per pin.  This is shown in comments below signal declarations section of source file, HEX421SH.TXT (normally .PDS, but can not post such extentions).  Then I will install current limiting resistors on my 110VAC Bridge Rectifier to limit it to 1mA max.  Then map out thermister characteristics to be sure it will resolve as needed (surplus unknown that works at about light bulb temps).

I see source JED downloaded only once.  If you want to use FF's AND, & OR gates, just follow the equations for the signals driving the 6 Prim's.  The logic is exactly the same.  You will see how many wires and chips this design saves.  Just ask..., else share what you think is better.  I really like Eldarian's Xylinx, but it just won't fit inside ring. 

Ward

[HumblePie]

@All,

Here is rough layout.  Details details...  I'm using LM193 single-supply Op-Amp oscillator circuit from the data sheet.  Duty cycle and range is better than 555 timer.  I will use DIP switches to select range Cap, extra LM193's to process Thermistor, Voltage Sens, and Feedback for HVPS (orderd GammaH'V' SM-10).  I need to add more shutoff inputs to the GAL.  The Kill-switch will share Gnd with DC power into terminal on right.  An LM7805 will provide 5V power to the GAL and 74AC14.  Unfortunately, I have little free time this weekend.

Ward

[HumblePie]

@All,

I need more room for Varistors, resistors, and more Cap's.  Maybe I can attatch those to the back side.  Maybe I'd better etch a PCB.

Here are some tests I did with MIC4427 at 15V driven by DDS 20 buffering gates;  CD40106 Vs 74HCT14 both driven at 5V.  I wanted to test MIC and UCC switcing response to fast Vs slow edge speeds.  The drivers do not seem to care.   I keep on replaying SM clues over and over about solid state ringing, soldering > 1/2" above PCB material, etc.and over and slap myself and say 'look at PWM3... it uses IRF540's with 51 & 39nS edge speeds, driven by OCP=PCP116 Opto's with 100 to 500nS edge speeds, and it still worked...  Maybe not with effects, but it worked at least in pulse mode.'

I suspect I may really want a slower driver like the TC4421CAT after all.  That way the ringing will be tamer.

Now Earl, how do I make my layout better?  I may be able to get driver pins to bend over to touch terminat posts.

Ward

[HumblePie]

@All,

I need advice about best choke found.  This photo is the best one I have found so far.  It is 0.6 H.  I will have to take my function generator to the surplus store if this is not low enough cutoff.  I stimulate one end of coil with 5V square wave and see what comes out other end.  This one lets only about .5V at 30KHz, just after resonance peak it drops off fast!.  Others were all > 100KHz or even higher.  I am new to chokes, but I see we want to pass as near to just DC as possible.  How low has anyone else found?  How you guys define "Cutoff" of the choke candidate?  I used point where I had only 10% of input amplitude, ignoring wave shape distortions.  Also attached is my xls showing firing order options inside GAL HEX421SH.PDS (txt).  I will make active low file soon so the hexFETs inverting will render correct pulse polarity for Non-Inverting drivers.

Other timing options require two GAL's; like 1/2/3X (requires 18X CLK source), 1/2/5X (30X CLK).  A five bit counter is needed Vs a 4 bit in this 1/2/4X GAL... no more outputs on a GAL22V10... the 10 is for # out's.  A 1/2/7X (42X CLK) timing GAL needs a 6 bit couter.  I plan on making GAL phase/pwm for defining start and stop postions like in ARM uCntrlrs, but at 200MHz for 5nS time increments... after all the other things I have to accomplish.  (no not DIP GAL's else too big PCB). 

Edit:  Well Eldarian's thread #57 Megger points out a better phase solution / pwm solution (2nd chip) in the DS1023-xxx programmable delay line.   Just that this needs a  bus to control it... I do dream big.

I have 1 gallon high-temp encapsulating epoxy I will use on next toroid so it will never buzz, much.  Shelf life remaining is about 11 months.  I am looking for suiable containers for sending small amounts to my lab partners in Texas.  Just need to arrange hazardous shipping.

So quiet.  It is OK to comment.  Please do.  Save me from mistakes.  Share ideas.  Please.  I am going to start messing with my ARM7024 for < 20KHz OK resolution to drive this thing that way too.

Ward

[Earl]

Hi Ward,

I assume this choke is to keep the radiant energy from shorting to ground via the HV supply.
Two comments:

1- you can always use a resistor is series with the choke.
2- I believe you could also use only a resistor to achieve the same isolation.  A 22 Megohm resistor should provide enough isolation.
My suggestion would be to use only several [chip ?] resistors in series.  The reasons being that if one should short or get too dusty or dirty or humid, you would still have electrocution protection in case of something stupid happening.  The current should be essentially zero, so resistance does not matter.  I would use a couple of 1M or 10M SMD chips in series.  The distance between total resistance ends should be sufficient to satisfy safety concerns of eventual arc-over.  An additional plus is that if the resistance drops any voltage, then you know you have a current leakage somewhere.  A series connection of resistors also drops any residual capacitance across the R, 3 Rs in series has 1/3 the capacity of just one resistor.

the choke shown is real nice, but very big.  A couple of series SMD resistors has much less volume and will fit into the center of the toroid.
Do not forget that such big chokes while effective at lower frequencies might present a short-circuit at higher frequencies due to inter-turn capacity.  There is no such thing as one choke being effective over a broad range of frequencies.

My gut feeling is that series resistors are preferable over an inductance for the following reasons:
1- current limiting against electrocution hazard
2- wider frequency range
3- much smaller volume

One time I had a 220V to 24 VDC switching power supply that had arcing problems because of too close spacing, especially in boat motor rooms (salt vapor in the air).  I coated the corresponding area with 5 min epoxy 2-component glue and that solved the problem.

Regards, Earl