Author |
Message |
slo
Joined: Feb 18, 2007 Posts: 204 Location: Victoria, BC
Audio files: 1
|
Posted: Mon Mar 03, 2008 9:46 pm Post subject:
Finished, but trouble with the Klee...... |
|
|
Well the Klee is finished, but it has bad behaviour.
Using the build manual, I ran it through it's testing paces but didn't get far. I noticed a stray gate bus 2 led on stage 5 with no gate 2 switch selected and in manual step stage 8-16 only light up on the even stages, 10, 12, 14, 16. Looking at the previous thread with the trouble with the CD4034, I switched these 2 IC's and now the Klee is totally screwed with the stage 1-8 Led's lit all the time.
When first powered and doing the testing, the Klee ran nice and sequencial with a clock signal in 16x1 mode, so I don't think there is too much wrong, probably these 4034? Wadda think? Oh, I triple checked the wiring.
Bill, I got the parts kit from you and the make of these 4034's is... well it says SCL on them.
George
Description: |
|
Filesize: |
39.48 KB |
Viewed: |
712 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
Description: |
|
Filesize: |
60.21 KB |
Viewed: |
704 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
|
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Tue Mar 04, 2008 8:11 am Post subject:
|
|
|
Hey Slo,
I'm ordering some CD4034s from Mouser today. Usually I get the order in the same week I sent it from them (sometimes even the next day). I'd like to put a couple of them in my Klee and test them to see if they work in mine, then I want to send the tested CD4034s to you to see if they work OK with you.
If possible, when you get them, and they fix the problem, I'd like to take a look at yours (IE, could you send them to me?). I'd like to study them
My built Klee has two TI CD4034s in it. The original breadboarded Klee has an RCA and an SCL in it. The built Klee is still ticking - the breadboard was still ticking when I took it off the bench a few months ago (it ran for well over a year).
PM me your snail addy, and when I get them tested, I'll send them on to BC. I should be able to get that done by Friday, if you don't mind the wait.
Take care,
Scott _________________ My Site |
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Tue Mar 04, 2008 8:16 am Post subject:
|
|
|
Forgot to mention - what a lovely panel, and it looks like nice work putting it together.
Second - does anybody know which manufacturer SCL is? I should know, but..I don't.
Cheerios,
Scott _________________ My Site |
|
Back to top
|
|
|
State Machine
Janitor
Joined: Apr 17, 2006 Posts: 2809 Location: New York
Audio files: 24
|
Posted: Tue Mar 04, 2008 10:26 am Post subject:
|
|
|
Let me also add that it seems that there has been trouble with these CD4034's and I believe it's the ones from me and not the fault of Scotts fine design. It seems a common thread lately that folks are replacing the CD4034's. The troubled IC's are the "SCL" ones that I got from Hosfelt Electronics. If any folks get any bad ones from me, I will gladly send you a pair of new ones free of any charges I am selling the Texas Instruments chips now and not these anymore. My TI's are working fine for several months. I am sorry for any inconvenience this may have caused anyone.
NOTE: The new parts order I am doing for this new batch of Klee orders will not use the SCL CD4034 IC's and I will functionally test each and every one I sell ....
Scott, if your TI's work in Slo's unit, then I will send you a pair for your trouble.
Bill |
|
Back to top
|
|
|
State Machine
Janitor
Joined: Apr 17, 2006 Posts: 2809 Location: New York
Audio files: 24
|
Posted: Tue Mar 04, 2008 10:31 am Post subject:
|
|
|
Quote: | Bill, I got the parts kit from you and the make of these 4034's is... well it says SCL on them.
George |
Yes, I am so sorry. Scott will hook you up then I will repay him in return for his parts. Also, NICE PANEL DUDE !!!!!!! Very nice work indeed !
Bill |
|
Back to top
|
|
|
State Machine
Janitor
Joined: Apr 17, 2006 Posts: 2809 Location: New York
Audio files: 24
|
Posted: Tue Mar 04, 2008 10:33 am Post subject:
|
|
|
SCL - Semiconductor Complex Ltd
Located in India I believe I think .........
Semi-Conductor Laboratory (SCL- formerly known as Semiconductor Complex Limited, established in 1983) is now a Society (registered on 08.11.2005 under the Societies Registration Act, 1860, as amended by Punjab Amendment Act, 1957), with the main objective to undertake, aid, promote, guide and co-ordinate Research & Development in the field of semiconductor technology, Micro Electro Mechanical Systems and process technologies relating to semiconductor processing.
SCL, through its in-house R&D efforts has developed 3 micron, 2 micron, 1.2 micron and 0.8 micron CMOS technologies as well specialized technologies such as EEPROM and CCD. SCL has over the years developed and supplied a number of key VLSIs, majority of which have been Application Specific Integrated Circuits (ASICs) for high reliability and industrial applications. |
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Tue Mar 04, 2008 11:32 am Post subject:
|
|
|
Ah, you don't have to repay me, Bill - you've helped me in so many areas to begin with. Besides, who was it that pointed you to Hosfelt to begin with?
That SCL part I have looks old (so does the RCA) - I wonder if they're a bit more susceptible to ESD, being older parts? But, then again, I seem to recall 23isgood having a TI go tits up on him as well.
In any event, I knew you'd step in and help here, so I figured I'd beat you to the punch, for a change.
I wish em had a field service budget - I could use a trip to BC. Anyway, everyone who bought boards will Klee successfully, or I'll die tryin'...
I can think of worse ways to go.....
Cheers,
Scott _________________ My Site |
|
Back to top
|
|
|
Luka
Joined: Jun 29, 2007 Posts: 1003 Location: Melb.
|
|
Back to top
|
|
|
State Machine
Janitor
Joined: Apr 17, 2006 Posts: 2809 Location: New York
Audio files: 24
|
Posted: Tue Mar 04, 2008 5:19 pm Post subject:
|
|
|
Quote: | Ah, you don't have to repay me, Bill - you've helped me in so many areas to begin with. Besides, who was it that pointed you to Hosfelt to begin with? Embarassed |
OK then, thanks Scott. I try ... I am sure Georges problem will get solved using the replacement shift registers Oh, my comment about Hosfelt, it's really not even Hosfelts fault. I don't want to blackball them. they have been a good supplier Thanks for the suggestion, they shipped my 70 of them right away !
Quote: | That SCL part I have looks old (so does the RCA) - I wonder if they're a bit more susceptible to ESD, being older parts? But, then again, I seem to recall 23isgood having a TI go tits up on him as well. |
Now that you mention, yes, the TI did go tits up I noticed that all the CD4034's are fairly oldish ... Are any fabs still producing them ??
Quote: | In any event, I knew you'd step in and help here, so I figured I'd beat you to the punch, for a change. Very Happy |
Oh, you did man !!! Yes, when I see people having troubles with circuits, sometimes I can't help myself
Quote: | I wish em had a field service budget - I could use a trip to BC. Anyway, everyone who bought boards will Klee successfully, or I'll die tryin'...
I can think of worse ways to go..... Laughing |
OK man, I will be right along side you dude !!!!
Bill |
|
Back to top
|
|
|
slo
Joined: Feb 18, 2007 Posts: 204 Location: Victoria, BC
Audio files: 1
|
Posted: Tue Mar 04, 2008 5:40 pm Post subject:
|
|
|
Hey, love you guys! Scott, Bill thanks so much for the Klee and all you have done to bring it to life, anytime you want come to BC let me know.
Scott, I will PM you my address and mail the wonky IC's to you. Bill, your parts kit made things so easy for me to build, a big thank you and no need to be sorry for something outside of your control.
George |
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Wed Mar 05, 2008 8:38 am Post subject:
|
|
|
TI still lists the CD4034 as active. They have two variants - the normal CD4034 and what appears to be a rather expensive MIL Spec CD4034. The "normal" one is available as PDIP, the MIL is CDIP, I believe. There are also surface mount varieties.
Maybe we should spec the MIL version
Cheerioios,
Scott _________________ My Site |
|
Back to top
|
|
|
23isgood
Joined: Nov 18, 2006 Posts: 236 Location: San Francisco, CA bay area
Audio files: 13
|
Posted: Wed Mar 05, 2008 11:37 pm Post subject:
|
|
|
Yes! Both brands went bad in my Klee. The first one to go bad was the SCL part. I just ordered two more TI parts just to be prepared. Though I still have one SCL that's still working fine.
pete _________________ Check out my music |
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Thu Mar 06, 2008 7:42 am Post subject:
|
|
|
I must live in the right spot - if I order from Mouser, I get the order the very next day, even with the cheapest delivery.
Anyway, last night I tested four CD4034s (TI) in Ye Olde Klee, and they Kleed OK.
Cheerio,
Scott _________________ My Site |
|
Back to top
|
|
|
State Machine
Janitor
Joined: Apr 17, 2006 Posts: 2809 Location: New York
Audio files: 24
|
Posted: Thu Mar 06, 2008 8:21 am Post subject:
|
|
|
Quote: | TI still lists the CD4034 as active. They have two variants - the normal CD4034 and what appears to be a rather expensive MIL Spec CD4034. The "normal" one is available as PDIP, the MIL is CDIP, I believe. There are also surface mount varieties. |
OK, thanks for that information Scott.
Wow, you get your Mouser stuff fast.
Bill |
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Thu Mar 06, 2008 8:33 am Post subject:
|
|
|
Yes, it's amazing - if I get the order in fairly early in the day, it more often than not shows up the next day. IIRC, Allied is just about as fast.
These CD4034s have a date code of 7BA38K8 - not sure how to decode that. They're different than the TIs in my Klee, though I failed to write those down.
Cheerio,
Scott _________________ My Site |
|
Back to top
|
|
|
slo
Joined: Feb 18, 2007 Posts: 204 Location: Victoria, BC
Audio files: 1
|
Posted: Wed Mar 19, 2008 7:44 pm Post subject:
|
|
|
OK Klee status.
Got the TI 4034's in place...thanks Scott.
Klee works as advertised with a few bugs.
Ironed out a few myself, 1 poor wire joint, 1 resistor out of place on the analogue board. (watch the extra holes)
Here are the ones left.
1. Still have a Gate 2 led, gate and trigger on stage 5 with gate switch set to gate 1.
2. In the Bus 1 reload test, leds only move to stage 3 and resets to 1 (should go to stage 4)
There might be more bugs as I move along in the testing/calibration, I'm pleased with so few in such a complex build and of course the ones I've found were my doing. It's real fun to hear that VCO singing a klee song.
George |
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Thu Mar 20, 2008 11:03 am Post subject:
|
|
|
OK. For whatever reason, your remaining bugs are associated with Gate Bus 1, obviously.
You can eliminate the CD4034s in this case - the same signals that are used to illuminate the pattern LEDs are also used to generate the gate bus signals. So, if your pattern LEDs are firing in the right order and loading when they're supposed to, etc., the encoder section is working fine.
If the majority of all the other Gate Bus functions are working as expected (every other gate/trigger pair works as expected with the gatebus switches), IC failure can be pretty much ruled out as well.
Having said that, at this time I can think of only one long shot scenario that may cause cause both problems - if the wires feeding gate bus switches 4 and 5 are reversed, you would see symptoms like that. But, I would think you would have spotted that, so I'll cover each one separately.
Stage 5 Gate Bus Switch Problem: Gate Bus 2 should only go high if neither gate bus 1 or gate bus 3 are high. Diodes D5 and D6 on page 4 of the digital board schematic form a logical NOR gate with with U11 of the digital board - in a nutshell, if a particular gate bus switch is sending an active high to either the gate bus 1 line or the gate bus 3 line, gate bus 2 uses the logic generated in this section to "stay low" and never generate a gate or trigger signal.
Diodes D5 and D6 must be fine, otherwise you'd see a lot more problems with gate bus 2; same goes for the logic down that line.
Each bit output from the CD4034s is sent from the digital board to the analog board. For the gate bus, these signals are sent through diodes D13 through D28 to the center pin of the gate bus switches. When a gate bus switch is in the center position, it will not send its signal onto either gate bus 1 or gate bus 3.
The diode feeding the register bit to gate bus stage 5 is D17 on the Analogue Board. The signal passes through it to pin 4 of J1 on the Analogue board, and from there to the center tap of the stage 5 switch. From there, the stage 5 switch should switch the signal either to gate bus 3 or gate bus 1. If it is not putting the voltage on either of those lines, then Gate Bus 2 will go high.
So, do this: Set up a one bit pattern with only the stage 5 pattern LED on. Check the voltage on the center pin of the stage5 gate bus switch - it should be high. Switch the stage 5 gate bus switch to gate bus 1. Check the voltage on gate bus 1, right at the gate bus 1 common connection of the switch. It should be high as well.
If the center pin of switch 5 is not high, check D17's anode to see if it is high and check its cathode to see if it's high.
Gate Bus 1 Load Switch - This is a problem of gate bus 1 going high when you don't want it to. If you are running a one bit pattern, and the pattern resets at stage 3 rather than stage 4, then gate bus 1 is going high when stage 4 becomes active. This would sort of hold with the off-chance that gate bus switches 4 and 5 might be reversed - if you think you're switching stage 5 gate bus high but stage 4 is actually switched in instead, gate bus 2 would stay on. And, if you have gate bus 5 high, but it actually is gate bus stage 4, you would get a reset after stage 3.
In any event, measure the bus 1 line (at the common connection between switches) and manually step a 1 bit pattern through with the test setup and look to see when gate bus 1 goes high. Note - do this with the stage 1 load switch OFF, otherwise once gate bus 1 goes high, it'll immediately flip low when the reset occurs, and you probably won't see it happen.
Let me know what you find!
Cheerio,
Scott _________________ My Site |
|
Back to top
|
|
|
slo
Joined: Feb 18, 2007 Posts: 204 Location: Victoria, BC
Audio files: 1
|
Posted: Fri Mar 21, 2008 9:48 am Post subject:
|
|
|
Scott, both problems fixed.
The gate bus 2 problem was a shoddy wire crimp @ the header on the analouge board for the switch.
The bus 1 reload problem was the 2 8-pin board interconnects, I had a housing on both jumpers reversed, pin 1-8 instead of 1-1, which made for some interesting gate sequences.
I have one more problem with calibration, I can't get -5v @ range 7, only -4.90v and my variable range only goes to -3.67v. I have it set up 5v and 4v in ranges 7 and Ext. with a 150k resistor in R33.
Mearsured 9.79v out of the 10v ref, is this ok or do I have a bad LM78L10?
thanx in advance
George |
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Fri Mar 21, 2008 10:24 am Post subject:
|
|
|
I'm not sure where my 78L10 landed, voltage-wise. That may be a tad low, but if the voltage is steady, I'd just adjust the resistor values instead.
The value of 10K for R32 (the divider resistor for range 7) is for an adjustable range of around 4V. If you replace R32 with a 22K resistor, you should be able to easily dial in 5V for range 7.
For my variable range, I just did a hail Mary with 150K and let that be the max level, whatever that came to (it was in the ballpark for me). I never did put in a way to precisely set that max range. If you replace the 150K with a 130K, that would put the max range in the neighborhood of 4.2 to 4.3V.
Cheerio,
Scott _________________ My Site |
|
Back to top
|
|
|
slo
Joined: Feb 18, 2007 Posts: 204 Location: Victoria, BC
Audio files: 1
|
Posted: Fri Mar 21, 2008 5:35 pm Post subject:
|
|
|
Tried the 22k resistor, but could only get 3v out of it, so looking @ the way the reistors drop for each stage, I would guess the value has to be lower than 10k for R32, to get a higher voltage. Uneducated speculation.
George |
|
Back to top
|
|
|
Scott Stites
Janitor
Joined: Dec 23, 2005 Posts: 4127 Location: Mount Hope, KS USA
Audio files: 96
|
Posted: Fri Mar 21, 2008 6:39 pm Post subject:
|
|
|
Better than my speculation. I got in a rush and was a-thinkin' backwards there. I'm very sorry about that! Of course a higher value is there going to drop more voltage.
Lessee.....
5.1K or 5.6K would work for dialing in 5V on range 7. Even if you had a 6.8K left over, that would probably work (highest range dialed in with it would be 5.8V, so that would get you to 5V).
Cheerio,
Scott _________________ My Site |
|
Back to top
|
|
|
slo
Joined: Feb 18, 2007 Posts: 204 Location: Victoria, BC
Audio files: 1
|
Posted: Sat Mar 22, 2008 12:29 pm Post subject:
|
|
|
Ok, got it dialed in. Went with the 5.1k so I have some room in the future if I want to nuts with a higher voltage. Klee is tested and and calibrated and working 100%, got to get my head around some of the functions, but I can see it's going to give much pleasure, it's deep!
Thanks to Scott and the crew, for the design and help.
George |
|
Back to top
|
|
|
|