Flow of electricity question: Active low

I'm building an interface board for this speakjet chip, that I've studied and it works good. I need to interface more pins from Arduino besides a basic serial pin. I tried the circuits on my own and I'm afraid to damage the pricey chip, and wait another 2 weeks to get another one, so I'm being cautious and getting confused.

Here is the basic diagram:

speakjet basic setup circuit

Pin B1 should be grounded, B2 & B3 should be high (5v) - for normal operation.

For my interface board I would like to tie the pins to Arduino digital output, to toggle different modes on the chip (such as change the baud for example).

--

It seems there's some type of strong signal on "B1 - MODE" pin. I had 10k resistor between Arduino and the B1 pin, but it seemed the Arduino signal was not strong enough to toggle the ground out of "DEMO MODE".

I tried connected B1 to ground using 10k 5.6k, 3.3k 1k resistor and the pin is still high and chip is still in demo mode. Seems like the only way for me to get out of demo mode is a hard wire to ground.

I'm just confused, should I just not use any resistor between Arduino and this pin - I thought that was kind of risky without any resistors.

The drawing you posted shows 3 "b1" pins, are you sure you aren't connecting to the headphone output "b1"? http://www.speechchips.com/downloads/speakjetusermanual.pdf

DocStein99:
It seems there’s some type of strong signal on “B1 - MODE” pin. I had 10k resistor between Arduino and the B1 pin, but it seemed the Arduino signal was not strong enough to toggle the ground out of “DEMO MODE”.

I tried connected B1 to ground using 10k 5.6k, 3.3k 1k resistor and the pin is still high and chip is still in demo mode. Seems like the only way for me to get out of demo mode is a hard wire to ground.

I’m just confused, should I just not use any resistor between Arduino and this pin - I thought that was kind of risky without any resistors.

I wouldn’t say it’s a “strong” signal on B1. You are not dropping the pin to a logic low with a resistor in there. Even with a 1k resistor, you are only passing 5mA to ground, which might not be enough to drag the voltage down. This isn’t a power source where you are trying to limit current. This is a logic circuit that you are trying to force down. Ditch the resistor.

756E6C: The drawing you posted shows 3 "b1" pins, are you sure you aren't connecting to the headphone output "b1"? http://www.speechchips.com/downloads/speakjetusermanual.pdf

I have the pins right and not confused with audio out. I have had the speakjet working for a while hardwired to test with. I'm just re-doing my protoboard so I can change the baud rate and change whatever else I can in setup mode.

tinman13kup: This isn't a power source where you are trying to limit current. This is a logic circuit that you are trying to force down. Ditch the resistor.

Ok, thanks. I'll try that, I was thinking to do that but didn't know if that was bad idea. The manual states is logic low signal.

DocStein99: Ok, thanks. I'll try that, I was thinking to do that but didn't know if that was bad idea. The manual states is logic low signal.

Stop. You definitely have something connected incorrectly. This is a cmos chip, any logic input can be easily pulled low through a 1k resistor. If the pin is not being pulled low then there is something wrong!

Unfortunately the linked datasheet is very light on electrical specifications for the inputs, however it does specifically say that the inputs can be tied either directly to gnd or through a resistor. Again unfortunately they don't mention any specifics on resistor values, however in this context 1k is already quite a low value.

I can summarize what I see on page 2 of PDF:

supply voltage: 2-5.5dc
supply current: < 5ma plus loads

outputs: 25ma

For normal operation reset must be connected to v+ through a resistor


M0, M1 mode select:

These pins are logical inputs that require a HIGH or LOW logic level. For normal operation, connect M0 to GND, and M1 to V+ through a resistor

I read manual many times, and can not find any other reference to resistance values. The rest of it explains how to communicate serial to drive the synth an speech.

The circuit I provided, is pretty much the only circuit that is actually provided in the datasheet for the Speakjet.

Link to Speakjet DATASHEET

I read manual many times, and can not find any other reference to resistance values. The rest of it explains how to communicate serial to drive the synth an speech.

Yeah they definitely mention using a resistor, but make no mention of the input current requirements. The 25 mA is the output capability.

When you say that you tried different values of resistor, did you happen to measure the pin voltages (with a voltmeter) for each different value? This will tell you how much current it (the resistor) is trying to sink to pull the input low and let you figure out what the input characteristic is like. From what you've posted however, it doesn't seem like any other cmos input that I've ever come across.

stuart0: Yeah they definitely mention using a resistor, but make no mention of the input current requirements. The 25 mA is the output capability.

A 1k ohm resistor? 100m ohm resistor? 2.2 ohm 20 watt non-inductive resistor? This crazy thing probably needs a photo/diode resistor.

stuart0: When you say that you tried different values of resistor, did you happen to measure the pin voltages (with a voltmeter) for each different value? This will tell you how much current it (the resistor) is trying to sink to pull the input low and let you figure out what the input characteristic is like. From what you've posted however, it doesn't seem like any other cmos input that I've ever come across.

Yes, with 10k I was getting like 1.5 or 1.7 volts, then I tried stepping down and to 1k and if I remember, was something like .8 volts (if I remember right). I stopped trying at 1k, I guess I can try down to 330-ohm, or whatever I can to get it to work (like put it away for another month and then come back).

DocStein99: Yes, with 10k I was getting like 1.5 or 1.7 volts, then I tried stepping down and to 1k and if I remember, was something like .8 volts (if I remember right). I stopped trying at 1k, I guess I can try down to 330-ohm, or whatever I can to get it to work (like put it away for another month and then come back).

Vcc is +5 volts, is that correct?

I would have though that 0.8 volts would have been a valid logic low already.

That input is sourcing 0.8 mA btw. With the very old (original) TTL you did need to sink (a max of) 1.6 mA to get an input low, though with later LS-TTL it was only 0.4 mA. Like I said, something seems wrong. That's not like any cmos logic input that I've ever seen.

If I remember right, the .8 volts with 1k resistor just about worked, except it flipped back into demo mode periodically and caused me to go over the rest of my circuits. At that point I got frustrated - and could have made mistakes somewhere with the 10 bypass wires I had to do on the solderless to try and troubleshoot what's happening. I wasn't making progress, which is why I posted here to get an outside thoughts.

I've also never seen a datasheet for an IC without a big confusing array of symbols, math and graphs on the first 4 pages. I am thankful it actually is written in a way I can understand.

I gather you're hooking up a Speakjet? If so, I hope the following "getting started" page will have some useful stuff for you... http://www.arunet.co.uk/tkboyd/ec/ec1speak.htm