Argh, should have mentioned what I've tried and checked so far:

• swapped IC112 (74HC595N--TRIG_HH) with IC113 (no change)
• swapped 4040N, 4011UB, 4013N, 74HC174, 4520N (no change)
• transistors were run through a component tester (before soldering)
• op-amps have been swapped with known good op-amps from my hi-tom (no change)
• no solder bridges
• no cold joints
• all pots, resistors and capacitors are the correct values and in the correct places
• continuity between main PCB HH output pin and I/O board HH input
• PROM contains correct data (verified using the Ride/Crash EPROM sockets)

Nobody? Okay, here's a more specific question, then.

Could someone with working hihats and an oscilloscope please tell me what sort of patterns should I be seeing from the 2SA798, 2SC2603s and 2SA115s?

Well I'm happy people are reading even if nobody's responding.

I've determined that all signals disappear somewhere between the 74HC174 and the resistor array. Narrowing that down is well beyond me so I'll try to borrow an expert to look at it. I'll report back.

I am pretty sure that your 2SA798 is defective. I experienced a similar issue. In my kits two of them didn't work correctly. The good thing was that I built two Nava's and could always compare measurements with each other. For example: on one board the ride was working and on the other one it was dead. So I measured both 2SA798 on the two boards in continuity mode with my DMM (I don't have an oscilloscope). The result were completely different values. I ordered new ones from China, swapped them and voila: it worked again. I could give you the values from my working transistor and you could compare them with yours. If the differences are too big, then it's probably broken.

Thanks for replying! For the record, the 2SA798s I have at Q78 and Q68 (crash/ride) seem to be passing voltage on pins 1, 2, 3 and 5.

The 2SA798 at Q84 (hi-hat) is passing voltage only on pins 1, 4, and 5. Pin 4 passes more voltage or less voltage depending on whether an open or closed or hihat is playing, so maybe it's only half-dead. :p

I would test it with a multimeter but unfortunately the data sheets I've found for the 2SA798 are all in Japanese. Which pins are which? :\

Maybe I should just gather the courage to desolder it and run it through a component tester. Aaargh.

KlangGenerator wrote: So I measured both 2SA798 on the two boards in continuity mode with my DMM (I don't have an oscilloscope). The result were completely different values.

The measurements I took from a known working 2SA798 (ride/crash) and the hi-hat 2SA798 were similar (680-700). I desoldered the hi-hat 2SA978 and ran it through a component tester. The results are in the same range as other transistors that I tested. :\ I was hoping for more dramatic results. I'll try buying a couple of spare 2SA798s anyway.

d-wreck wrote:I'll try buying a couple of spare 2SA798s anyway.

The schematic for the hi-hat shows the 2SA798 quite a ways after the resistor array, but what do I know about anything. The 2SA798s arrived this week. I swapped them in but neither of them made a difference. :\

The poor Nava build documentation, the complete lack of testing of the BOM, the board fixes that showed up after I'd soldered everything, my terrible skills, the time, the money, and (most frustrating) my case mashing down the stupid cheap square tact switches rendering the whole thing unusable... I should have just bought a plugin. Into the closet it goes.

Regardless, I appreciate everyone's advice.

hey dwreck, liked your blog.

Now that I finished my Nava I can try to help you, mine unfortunetly for you worked fine from first shot on, but this isnt my first project and so on

I can probe around to see what is missing in your circuit, but you have to tell me what you miss or what you wanna know.

Basically the hihat comes from a sample, with all the old logic circuitry involved, but i guess you already knew that.

So what could cause the click, is maybe a wrong timing circuit, this involves the inverteres and flipflops with some resistors and caps for the timing, should be 60kHz, if this is way too fast it might give a simple click.

Also some diodes could be wrong those determin the end address of the samples I think, if the ending address comes too soon maybe only half a sample or no sample will be played ( that is my assumption) I got no clue if that can be true, but this is the process of troubleshooting.

First of all always check the power supply of the hihat, the pins +15V -15V +5V and so so, those should always read the values , if its missing, then something else is terribly wrong.

I had have fucked up 27c256 once. Try erase it and program again with proper sample.