Tape recorder project

[EmilFrid]

I had an old Fostex G16 that got messed up and unusable after an unobtainum microcontroller in the remote failed. No problem, we found a much better machine, 1".

Now, I'm turning the old fostex into a really cool 4-channel 1/2" machine. It's great for singer-songwriters who don't need a lot of tracks and don't want to pay for unused real-estate, so to speak, and it's a great learning journey for me.

So I'm only keeping the bias oscillator and the transport from the fostex, everything else is being designed in the workshop. This far, I've finished a playback channel and the record channel is as good as done too. I'm going for IEC equalization. I've also made a controller and the shell of a meter bridge, as well ass re-painting the chassis. I've breadboarded the digital circuitry for the rec and safe/ready functions. I just need to turn it into a pcb eventually.

And here's the question for those smarter than I: the tape counter! How would you go about designing one of those? My idea is to use a simple 4 digit 7 segment type display, and I don't really have to be able to store cues like on our other machine. Just a fairly accurate counter.

Fortunately there's already an encoder under one of the rollers in the tape path (as there normally is) and the associated circuitry is working, as the syscon card is intact and healthy. But that's where it ends. I'm left with nothing but a signal to feed a counter circuit which is gone because the controller is kaputt. And the encoder sent it's information to the broken microprocessor in the controller via some serial data arrangement, which gives me no clues from reading the schematics.

I've looked around online but a lot of the stuff I find is confusing - I'm having no problems with audio electronics, and the archaic digital circuitry I designed for record and safe/ready functions was an easier task than expected, but im kind of stuck on the tape counter situation.


If there's interest on here, I could also make a few updates along the way.

 

[john12ax7]

Can you probe the encoder data to see what's coming out? You could maybe adapt the data to feed MTC circuitry.

 

[EmilFrid]

@john12ax7 I'll take a look at that when I return to the workshop. Maybe we can find a solution with MTC. Thank you!

 

[john12ax7]

MTC = Midi Time Code
I would assume there are some off the shelf chips or standard circuits to interpret the bytes and then drive a display that might be useful or insightful.

 

[EmilFrid]

@john12ax7 yes, I realized that. My brain is a bit out of order today.

Ok, interesting. So I guess we'll be looking for, or perhaps looking to generate, a varying square wave in that case? I'm pretty sure that's what the encoder puts out at the far end of its circuitry already.

 

[john12ax7]

This is a lot of speculation on my part, but when you say encoder and associated circuitry is working, I'm thinking that is outputting a digital signal of 1s and 0s. If you can hook that up to a logic analyzer and then engage the transport controls, it should give you an idea of the format of the data.

 

[EmilFrid]

I see. Will check, and I'll dig out the schematics for this part of the circuit.

 

[EmilFrid]

Here it is. Top part of the schematic

 

[gyraf]

..the pulses at U101 pins 6 and 8 should be directly usable for feeding into an arduino set up as quadrature counter. In these modern risc-computer times it's relatively simple to implement locating functions. I would not aim for external sync'ing though - mechanical timeconstants in tape recorders are generally a nightmare to try to map

/Jakob E.

 

[ruffrecords]

The way these normally work is the sensor outputs two signal in phase quadrature (90 degrees out of phase with each other). This allows the logic not only to count but to know which way to count. You can make a fairly simple piece of logic to convert the quadrature signals into a count and up/down signal. Or you could just feed them into a little 8 bit mIcro and do that in software and also drive the seven seg LEDs.

Edit: +1 to using an Arduino

Cheers

Ian

 

[EmilFrid]

@gyraf @ruffrecords thank you for the explanations, as I'm embarrassingly bad with digital stuff. But it's a learning journey.

I'll try to navigate the arduino jungle and see if I can find a feasible way of implementing it in the project. I'm afraid it will be a pain in the ass using code though, if that's needed.

 

[EmilFrid]

I've been looking at some arduino based circuits, but boy am I stuck in the past... I'm wondering if it might be better to seek out other alternatives. Sure, more soldering and more physical work, but also more hands-on and no coding or software which I for some reason never got along with (I'm writing this message on a clay tablet, and then I have someone at Microsoft turning it into digital text for the forum).

I've looked at some IC:s that might be interesting. For example, the ICM7217. Attached image is from the datasheet's "typical" circuit. Since I'm not very familiar with this type of circuits, I'm still not sure whether it would actually work. It has 4 BCD inputs and a "count" input. The up/down has to be manually switched, but it wouldn't be a big deal to just let rewind trigger "down" and play/fast forward trigger "up".

If this is a horrible alternative, let me know. I'm grateful for all the feedback I can get.

 

[JohnRoberts]

I still have scars from my time managing a 4 Track cassette product for Peavey last century. My first thought is why?

I will add to the chorus, that a modern MCU is probably the best solution, while it is likely learning a new skill set. FWIW coding MCUs may offer other benefits down the road for other projects.

JR

 

[EmilFrid]

I still have scars from my time managing a 4 Track cassette product for Peavey last century. My first thought is why?

I love tape machines. Why not?

I will add to the chorus, that a modern MCU is probably the best solution, while it is likely learning a new skill set. FWIW coding MCUs may offer other benefits down the road for other projects.

JR

I guess so. I'll keep that in mind, while still seeking out other alternatives.

 

[john12ax7]

Is this what is going on with the encoder?

https://en.m.wikipedia.org/wiki/Incremental_encoder#Quadrature_outputs

 

[EmilFrid]

@john12ax7 yes, that's correct!

 

[EmilFrid]

Ok, so it looks like maybe I could handle a basic Arduino based system.

I already found a program for the AtTiny85 chip reading quadrature and driving a 4-digit 7-segment display, so I won't have to code it myself, but perhaps modify the code a bit. I need the display to actually count time. If I've understood this correctly, I only need to get an Uno to program the chip.

Interesting!

In other news, the first record channel was finished last night, and I'm halfway through soldering the logic for the safe/ready functions. I should probably post some pics on here eventually.

 

[ruffrecords]

Ok, so it looks like maybe I could handle a basic Arduino based system.

I already found a program for the AtTiny85 chip reading quadrature and driving a 4-digit 7-segment display, so I won't have to code it myself, but perhaps modify the code a bit. I need the display to actually count time. If I've understood this correctly, I only need to get an Uno to program the chip.

Interesting!

In other news, the first record channel was finished last night, and I'm halfway through soldering the logic for the safe/ready functions. I should probably post some pics on here eventually.

The ATtiny range are exceptional; it is amazing what you can achieve with 8 pins!. (The 85A seems to be in short supply at the moment). The 84A is a 14 pin device with a dozen IO lines. I am using one right now for a VU LED meter project.

You are right about the Arduino. You only need it to program the chip you want to use. You can get low cost Arduino shields that will program the whole ATtiny range. I use this one:

DIY Arduino shield Attiny programming AVR Programmer - FREE SOP8 Adapter | eBay

There are cheaper ones but they don't have ZIF sockets.

Cheers

Ian

 

[EmilFrid]

@ruffrecords yes, quite interesting. I'm thinking rhis might be a good introduction to arduino.

I too noticed that the 85 seems to be in short supply, but I bought the last one at a Swedish web shop, hehe...

Thank you for providing some info.
If I was any good with coding I would've tried arduino for the safe/ready and record functions as well. But I'm making 4000-series logic for these tasks now.

After the millionth time I still forget the hassle of turning a breadboard circuit into a prototype board...

 

[EmilFrid]

So today my Uno and the serial display arrived (as well as an Attiny85 not in picture). I will read up on how to set this up, but if anyone on here with experience wants to come with some What Absolutely Not To Do's or tips and tricks, that's more than welcome. I guess first step is to download software...

Until then I'll finish up the safe/ready logic, which I've revised a couple of times, and try the rec and reproduce cards as part of that logic for the first time. I'll eventually make some measurements of frequency response in REW. Also have a feeling it'll take some adjustments before I'll get a nice punch in and out.

After that I'll start on the VU buffers.