Problem with Inchworm ICD

[jeg223]

I recently purchased the Inchworm kit from dipmicro.com. When I try to connect to MPLAB I get :
ICD00warn54 MPLAB has lost communication with ICD 2
ICD0021 Unable to connect with MPLAB ICD2
I get the same message when I try to download the OS. I disabled the FIFO buffers and set the flow control to hardware and I'm using a straight through serial cable. I've tried two different computers and reinstalled MPLAB v7.52.
The kit came with a .22:mu: F cap for the output filter on the 7805 so I replaced it with a 47:mu: F 10V and I made sure the crystal was off the board.
The power LED lights and the voltages at the test points are good. I'm using a 12V 1200mA wall adapter.
Does anyone have any suggestions? I might try changing the transistors. Can I use 2n2222 or do they have to be BC548?

 

[blueroomelectronics]

The transistors will not affect communication, don't change them. They are CBE types and the 2N2222 is a EBC type...

A few dipmicro units had bad crystals, if you can get your hands on a 20MHz computer type crystal try that first.

The caps are non critical, anything from 2.2uf thru 10uf should work. If you got the 12-13 VDC on D6 then all is well with the caps. The 12V is fine, it'll keep the 7805 nice and warm.

So far most problems are.
Cold solder joints
Bad crystal (only dipmicro units so far)
Poor quality USB to RS232 adapters (I recommend units that use the PL2303 with drivers from my site)

 

[jeg223]

Dipmicro said all their crystals are tested. Is there any way to test it in the circuit? I'm using a serial cable, so I guess the other option is to check my soldering.

Thanks for your quick reply.


Jim

 

[gregmcc]

See if this helps - I was having a similiar problem. Also see the loopback test from 3V0

https://www.electro-tech-online.com/threads/inchworm-components.26878/

 

[jeg223]

My Inchworm failed the loopback test. I double checked the continuity of the serial cable to make sure its not a null modem. The port communicates with my basic stamp.

Are there any components in the communication circuit besides the chips that could fail, or is it most likely a soldering problem?

Thanks for the tip, at least that narrowed it down for me.

Jim

 

[3v0]

My Inchworm failed the loopback test. I double checked the continuity of the serial cable to make sure its not a null modem. The port communicates with my basic stamp.

Are there any components in the communication circuit besides the chips that could fail, or is it most likely a soldering problem?

Thanks for the tip, at least that narrowed it down for me.

Jim

Look at the schematic. IIRC only the MAX232 and its caps are required to make it work.

 

[jeg223]

Could it be a bad chip then? The diodes and caps tested good (in circuit) but the loop test failed. I tested the continuity between the max232 and the pic and the connector.

The power led lights dimly when I connect to hyperterminal. I get +10V on R1in (RTS) and -10V on R2in (TxD) with the power connected.

 

[3v0]

This may sound stupid.
The inchworm needs power via the round connector.
The dim power led with the serial cable attached tells me it is not getting power.
Either you have not hooked up the power or you have a bad wallwart. (the tranformer adapter thingy)
If you have it powered check +5 between the gnd and +5 supply pins on the MAX232 or PIC.

 

[Omar.M]

Hello,
I don't have much experience with the Inchworm myself (glad to have Bill around on these forums, eh?), but it seems that Dipmicro Electronics had some problems with the crystals they were supplying with the kits (as Bill suggested). Just wanted to give you a heads up if you weren't aware of it

After reading their front page, this is what it says:

Those who purchased Inchworm before April 5 are eligible for free 20MHz crystal
Please claim the crystal in customer notes on your next order. Ends June 30, 2007.

So, I do not know when you have bought your kit, but if it is the crystal problem, then you can get them to send you the 20mHZ crystal.

-Omar

 

[jeg223]

Its not stupid, I'll take any help I can get.

My voltages when powered are +5 at the test point D7, VCC on the max232 and at VDD on the Pic. The voltage at the VPP test point is 13.5 with no cable and 12.9V with the cable in.

I mentioned the dim led because it means "something" is coming through the serial cable.

Here's where I am right now:
Communication error when I try to connect to MPLAB or download the os for the 16F877
Voltages test good at the test points and at VDD
Serial cable is straight through and computer communicates with the Basic Stamp
Caps and Diodes around the max232 tested good.
Circuit is closed between the DE9 connector the max232 and the Pic, no visible solder bridges or bad joints.
The chip is oriented correctly with the notch to the top of the board.

Inchworm failed the loop back test

Maybe I'll set a gremlin trap and see if anything bites.

 

[3v0]

The power led lights dimly when I connect to hyperterminal. I get +10V on R1in (RTS) and -10V on R2in (TxD) with the power connected.

The power light should come on full when you hook up the ac adaptor.
The busy light is controlled by the PIC. You should not see any activity on it when doing a loopback test.

 

[3v0]

You have me confused.

I have talked about this a few times.
Tell me exactly what you did for a loopback test.

 

[jeg223]

I pulled the PIC out of the socket
connected the serial cable, started hyperterminal
set to com 1...That's when the power led came on.
connected the AC adapter (Brighter power LED)
Using a small jumper, I connected pin 25 to 26 (RX and TX) of the PIC socket.
I hit a few keys on the keyboard, but nothing happened in the terminal window.

 

[3v0]

I pulled the PIC out of the socket
connected the serial cable, started hyperterminal
set to com 1...That's when the power led came on.
connected the AC adapter (Brighter power LED)
Using a small jumper, I connected pin 25 to 26 (RX and TX) of the PIC socket.
I hit a few keys on the keyboard, but nothing happened in the terminal window.

Did you have hyperterm set to "no flow control"? I know you tested it with the basic stamp but humor me.

Start from the cable and work you way over to the PIC. Use hypertem to do a loopback in each of the following steps. I understand that step 1 will pass and step 4 will fail in your case.

1. Remove the RS232 cable from the Inchworm and short pins 2 and 3 on the cable.

2. If step 1 passed: Remove the MAX232. Look at the schematic and determine which pins on the MAX232 socket connect to 2 and 3 on the DB9. Short these pins at the MAX232 socket. If this works you know the RS232 signal is getting to the MAX232. If not you have bad solder joints between the DB9 and the MAX232 socket.

3. If step 2 passed: Replace the MAX232 and remove the PIC. Short the MAX232 pins that go to TX and RX on the PIC at the MAX232 chip. If this works your MAX232 is working.

4. If step 3 passed: Short the TX and RX pins on the PIC. If this fails you have bad solder joints between the MAX232 and the PIC.

 

[jeg223]

I had the flow control on hyperterm set to hardware. I set it to no flow control and all the tests passed. Does this mean the PIC is bad or unprogrammed.
Or its a bad crystal.

Thanks for the help, your troubleshooting is top notch

 

[3v0]

Ignore this.

I said to ignore this.

 

[3v0]

Could be either. Do you have a scope or a logic probe ?

Give this a try first. Maybe you did it but try again.
In MPLAB set the device to PIC16F877A.
Then download the OS.

If you have a scope with a 10:1 probe you should be able to see the crystal in action. Do some checking on the old threads to see what was said. Only works on one of the crystal pins. The other stops the crystal.

If you do not have a scope a logic probe with a high enough input impediance might show pules if the crystal is working. IIRC the Mondo superProbe will check the frequency. But I like the scope better because you can see the waveform.

 

[jeg223]

I don't have a scope or a frequency counter with the bandwidth to check the crystal. Is there any indirect way of testing it? I have a replacement crystal coming next week so I'll let you know if that works.

Thanks for your help

 

[blueroomelectronics]

If you do not have a scope a logic probe with a high enough input impediance might show pules if the crystal is working. IIRC the Mondo superProbe will check the frequency. But I like the scope better because you can see the waveform.

Did you actually build the Mondo superProbe?

PS hope the crystal works out to be the problem. The loopback test should have worked though regardless of the crystal.

 

[3v0]

He has the loopback working all the way to the processor TX RX pins. I was thinking he just needed to download the OS but that did not fix it.

I have built two version of the Mondo SuperProbe. The first has the processor and display stacked one board on the other. The second and more useable one is a single board. I have Eagle files for both and could post them.

I used a resonator instead of a crystal. According to the author of the circuit the resonator is not stable enough for the video signal generator but all else works. It should not be too hard to add the caps so a crystal can be used. Added a bypass cap for the PIC.