• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Question re: grounding an IC with another IC

Status
Not open for further replies.

Tickled_Pink

New Member
I'm currently fiddling about with the MAX7300 I/O expander. One of my ideas is to use it as an interface to my Atari 8-bit (again) so that it can be used as a digital IC tester. In order to do this, I need to be able to manipulate all the pins on the ZIF socket into which the IC to be tested will be inserted.

Obviously, not all ICs have the same number of pins or even pin arrangements, so I'm hoping that the 7300 can be used to provide both power (Vcc) and ground for the numerous ICs.

My question is concerning the grounding of the target IC. Basically, can one IC's input pin be used to ground another IC through the target's GND pin. If so, are there any issues I might need to be made aware of?

Thanks
 

Roff

Well-Known Member
The MAX7300 can only sink 10ma (this would be for target ground pins) and can source 4.5ma (for VCC pins). So, the answer is no, you can't use the MAX7300 to directly apply power to another IC. Maybe you could use relays (reed or other) to apply power as needed. You would configure the MAX7300 pins connected to target power pins (and relay contacts) as inputs. It does seem like you will have a relay packing density problem if you do this, unless you limit the relays to the most commonly used power pins.
The inductance of the relays (not the coils, the contact path) may be a problem for high-speed target chips.

Ron
 

Tickled_Pink

New Member
Not being exactly an expert on relays, I'd appreciate it if you could clarify for me what they actually do. I take it from looking at the catalogues that they're electronic switches?

Unfortunately there are rather too many ... any suggestions as to the type (or specific model) which might do the trick for me?
 

kinjalgp

Active Member
Relay in not a good soultion for such application. Use CMOS switches like CD4066 or ADG411.
 

Roff

Well-Known Member
kinjalgp said:
Relay in not a good soultion for such application. Use CMOS switches like CD4066 or ADG411.
CMOS switches have too much resistance and capacitance to be used for switching power to some ICs. I agree that you could probably use them for some devices. You could probably also use a bipolar transistor for many applications. When I suggested relays, I was thinking of a general purpose IC tester for ICs that may draw as much as 100ma, or even more.
I posted a simple relay driver on the projects forum for Mr. Pink. You may opt for one of the other solutions if your target ICs are low power.

Ron
 

kinjalgp

Active Member
ADG411 is a precision analog switch with on resistance below 35 Ohms which is comparable to relay's contact resistance of 10-15 Ohms.
Switching speed of 411 is 175nS and relay comes no where near it (20-30mS).

I have seen IC programmers which use ADG411 for Vpp and Vcc switching.
 

Roff

Well-Known Member
kinjalgp said:
ADG411 is a precision analog switch with on resistance below 35 Ohms which is comparable to relay's contact resistance of 10-15 Ohms.
Switching speed of 411 is 175nS and relay comes no where near it (20-30mS).

I have seen IC programmers which use ADG411 for Vpp and Vcc switching.
Well, I don't want to go overboard promoting relays, because I admit that other options may work for our OP, and I don't want to get into a pissing contest with kinjalgp, but I hate to see questionable information go unchallenged:

1. Most electromechanical relays have contact resistance in the 100 milliohm range. The one I suggested has 100 milliohms MAX contact resistance. I would like to see a spec sheet on a relay with 10 ohms contact resistance.

2. 35 ohms is probably fine for Vpp (programming voltage on an EPROM), but if I were building a general purpose IC tester, I wouldn't want that much resistance in series with either power pin - certainly not in the ground lead.

3. Most small relays switch much faster than 20ms. The one I suggested, http://oeiwcsnts1.omron.com/pdfcata...DE56D0286256D3500512F89/$FILE/D20G5V10503.pdf, has 5ms maximum switcing time plus 5ms max bounce time. Typical times are much faster. Switching time really isn't important for this application anyway. The relays apply power to the appropriate pins, and they stay on until the test is over. The time to insert and remove the IC is going to dominate, unless the tester is in an automated setup.

Ron
 

kinjalgp

Active Member
What you say might be correct.

But one quesion: How do universal programmers switch VCC, VPP and Ground between different pins. I am sure they don't use 40 Relays for a 40-pin socket. They use something called "Universal Pin Drivers". And I guess these are MOSFET switches which has some resistance. But you say resistance in ground is not good. So how actually is it done?
 

Roff

Well-Known Member
Good points, kinjalgp. I found a couple of pin drivers by doing a Google search. Elantec has one for "low" speeds (to 10 MHz!) that looks pretty good: http://www.elantec.com/pages/pdf/EL7154C.pdf. It looks like you could use this to apply power or logic signals to a pin, so long as you don't need the logic levels to be different from the supply voltages. This would be fine for functional testing, which is probably what our OP wants. I don't think it would work for testing clocking speed limits or noise margins. Analog Devices has a blazing 500 MHz part which is apparently designed for testing high speed devices, but I'll bet it costs a fortune. It also has higher output resistance, which could be a problem for some devices.

Ron
 

kinjalgp

Active Member
Well, I had already looked up to that IC while seacrhing for what "Universal Pin Drivers" mean but still I think I got one problem. it has two MOSFETS for switching between VH and VL. But what about Vpp i.e the Programming Voltage. I am looking for a chip which can switch between all these three voltages.

The reason why I am asking this is I want to build a Very Low Cost Universal Programmer so that all hobbysist can afford it who want to try their hands on different microcontrollers and PLDs. The commercially available universal programmers are just out of reach of most people. It will be an open design. If anyone likes to join do let me know. I have not yet done any work on it but if I get good response from you people, we can start the work.

I have also searched for any open design Programmers and have also found some. But I am not satisfied with their devleopment. Some are totally on the wrong track and some are trying to build a programmer which will have different adpters for different devices (Just imagine, you are going to build 8000-10,000 device adapters for making it universal). In this case I would prefer to buy commercial one rather than building it.

I want to build it such that it does not require any adapater upto 40-pin devices or may be say 48 pins.
 

Roff

Well-Known Member
Won't you at least need a lot of different sockets to accommodate the large variety of package styles?
 

kinjalgp

Active Member
Yes Package adapater will be required. But no adpaters for DIP devices. It should accomodate all DIP devices upto 40 or 48 pins.
 
Status
Not open for further replies.

Latest threads

EE World Online Articles

Loading
Top