Intermountain F7B DCC won't move


L

LotoQuestions

Member
Intermountain F7B DCC won't move

Urgencey has caused me to skip the introductory message and hop right to the questions, back to intro later.

Intermountain F7A & B factory sound & DCC from Ebay, Digitrak Zephyr, 6' test track

Work fine, set loco addr to match unit#s, consist them and seems fine.
Both units work, speed matches, couple and all seems fine.
I think that I lost track power at some time - loose connection.

Later I notice the B unit is not powering up, just dragging.
Remove consist - still wont move - makes engine noise but doesnt respond to Fkeys.
Reset by setting CV49=110 CV50=255 CV56=113 ???


Still wont move even with Analog, some sounds work.

Suggestions??

thanks ,,, ken
 
Last edited by a moderator:
open it up to check to make sure nothing is binding.

Then, try resetting again. I am not sure which sound system Intermountain uses, but if it is QSI, you'll have to send it back to the factory. I've seen a couple QSI units do that (just sit there and nothing happens but the sound revs up).

Try resetting, then program the decoder address again. Also check to make sure that CV2 and CV3 are set to 0 (CV2 is start speed, CV3 is start momentum) then try again.
 
More info

Sounds still work (or started again) when using the Loco addr I programmed. For instance F2 sounds horn.

Therefore I have not properly reset the decoder.

Decoder is at least partially functional since I can read and write CVs for example CV2 & 3 from above note .... thanks.

I dont think it is a binding issue but will dissassemble once I get a proper reset if I have no luck.

Wondering about the limited docs I have.

thanks ,,, ken
 
your unit has a QSI sound decoder. If none of that works to fix it, you'll have to send it to intermountain for repair.

Those QSI decoders are terrible
 
Solved .....

As in the previous note I could not get the decoder reset no matter how many manuals I referred to and values I plugged into CVs.

In the Quantum 272 page heavy doc they refer to a magnetic reed switch for master volume control and reset.

Following diBurning's advice, I removed the shell to check for binding in drive train and to see what version of the QSI decoder I had.

Found no binding but did see that I had the later version decoder with reed switch as illustrated in the QSI manual on page 260.

Having no supplied wand I used a small magnet and got a response on the volume control. I then followed the reset instructions and magically no longer a response on the unit # --- it is back to loco 3 - must be reset and yes it moves.

Glad that the magic smoke had not escaped and thanks to diburning for prompting me to open it up. I wasnt looking forward to seeing a burnt decoder or motor on my new toy.

thanks all ,,, ken
 
The wand is just a magnet so any magnet will work. As you've found, QSI decoders are almost impossible to reset from a Zephyr. Note the position of the reed switch and you can use the magnet right above it with the shell still on to change the volume settings or reset the decoder. I've gone through this drill several times with my P2K GP-7 with a QSI 1a version decoder.
 
Saga Continues

Well the decoder is reset but now I notice a problem that I dont think was there before.

Unit surges badly in middle half of speed range, not always to the same extent. Almost seems like a loss of elec contact.

Occasional miss of a beat at lower speed but hard to tell - I am impressed with the low speed performance.

No noticeable problem at top speed -not very high for these units.

Before the decoder problem this B unit was run consisted 6" away from the A unit and I am quite sure that I would have noticed a problem but it is possible I didn't run at mid speed.

Can this be something in the decoder due to the reset?
Or decoder sensetivity to electrical- ie dirty track / wheels?
Or time for a general forum?

Can't see how I can have screwed anything up taking the shell off.
If it is mechanical, is there a common problem with Intermountain HO?
Is breakin likely to help.

I am puzzled because I would have thought mechanical or electrical would show up consistently across the speed range.

thanks for listening ,,, ken

Cant wait for the locobuffer to be delivered!
 
My friend has an engine with a QSI decoder that is doing the same thing. You WILL need to send it back to the manufacturer!
 
Decoder or??

diburning said:
My friend has an engine with a QSI decoder that is doing the same thing. You WILL need to send it back to the manufacturer!
Click to expand...

Thanks for the info. This at least goes part way to reasuring me that this latest fault is an artifact of the decoder mishap and the attempted and then successful reset and not some major mechanical fault.

How has your friend ruled out problems other than decoder?

thanks ,,, ken
 
As far as I know, the only three-step, three-CV, reset is with QSI decoders, and you have indicated that you are dealing with a Digitrax.

There-in lies a problem. Try CV8 set to "8".

-Crandell
 
QSI and Zephyr

Selector said:
As far as I know, the only three-step, three-CV, reset is with QSI decoders, and you have indicated that you are dealing with a Digitrax.

There-in lies a problem. Try CV8 set to "8".

-Crandell
Click to expand...

The decoder is a QSI in an Intermountain with Digitrak Zephyr as the DCC base. Hard reset with the magnetic reed switch worked as far as reset goes.

thanks ,,, ken
 
Okay, I'm straightened out now. :p

Could you describe more about the mid-range surging? Does it accelerate normally under a 70-80% CV3 setting if you dial in about 40 on the speed step indicator or throttle knob? The way a decoder should work if the speed steps are set to 28/128 (have you verified this...?) and you have something like 70-80% of the maximum allowable inertia programmed into CV3, is that when you snap the throttle quickly up to a speed step near 40-50, the model should accelerate more or less consistently until it reaches the set speed. If it doesn't, and surges quite badly near speed step 20 or so, then you might need to set CV6 down quite a bit since it is the mid-range voltage setting for 128 speed steps. I would find out the permissible range for programming values into CV6 for your version of QSI, and maybe half that, or at most about 65%. Exit programming, and repeat the quick throttle-up to speed step 40 to see if the acceleration smooths out noticeably.

It is a quick solution if it works for you.

-Crandell
 
Last edited by a moderator:
It's a speed table mismatch that happened after the reset. Had the same thing happen to me. Enter 2 in CV 29 and that should restore the correct 28/128 speed step table and the engine will operate smoothly.

DL, I don't know why you keep saying it has to be sent back to QSI. I've reset my QSI decoders a number of times and, except for the speed table glitch, it has performed fine. There are issues with the QSI deoders that don't have the magnetic reed switch to reset the decoder but even most of those can be handled with the three step procedure Crandell referred to. I have three engines with QSI decoders and have never had to send one back, especially if you have Decoder Pro. That will handle almost any QSI problem.
 
Surging better description

Thanks for all the suggestions. Detail work will have to wait for the grass to get cut - short break now but later tonight......

Descriptions are the hardest part and things that seemed obvious visually dont seem so cut and dried after rereading my notes and thinking things out.

Surging happens repeatedly - say every Few inches or so - but not totally consistent and only in the middle speed ranges. Definitely not just once at reaching a certain speed - perhaps oscillating between the srange jumps.

Almost as if it looses contact momentarily, just the flywheel action keeping it going for an inch or so and then continues with what seems like a rush. Horn will sound consistently with no audible breaks.

Definitely runs clean at highest speeds.

Still quite sure? that this did not take place before the "won't move" issue.

Dont really want to use the A unit, even to check it's CVs - am trying to wait for the LocoBuffer so that I can save the A configuration in case I have an issue with it as well as I still don't know what happened in the first place.

Eager to try the suggestions tonight.

thanks ,,, ken
 
Last edited by a moderator:
UP2CSX said:
It's a speed table mismatch that happened after the reset. Had the same thing happen to me. Enter 2 in CV 29 and that should restore the correct 28/128 speed step table and the engine will operate smoothly.

DL, I don't know why you keep saying it has to be sent back to QSI. I've reset my QSI decoders a number of times and, except for the speed table glitch, it has performed fine. There are issues with the QSI deoders that don't have the magnetic reed switch to reset the decoder but even most of those can be handled with the three step procedure Crandell referred to. I have three engines with QSI decoders and have never had to send one back, especially if you have Decoder Pro. That will handle almost any QSI problem.
Click to expand...

Not everyone has decoder pro or a computer to decoder interface. My friend ahd to send his back because it was doing what he described. Atlas replaced the decoder.
 
DL, I think your friend got some bad advice unless the decoder was truly defective, which I doubt. I suspect Atlas just reset it to the correct speed step table and sent it back. Decoder Pro works even if you have no computer connection to the DCC controller. It can run in Loconet simulator mode. You just enter what changes you want to make, go the CV sheet, see what's changed, and enter those new values in the correct CV's. Assuming he already has a computer, JMRI, which includes Decoder Pro, is free, so you can't get cheaper than that. Certainly, entering 2 in CV29 does not take a rocket scientist to do.
 
Last edited by a moderator:
Status update - Intermountain

Thanks for the suggestions so far. Was waiting for the LocoBuffer to arrive but it has not answered my questions about this problem.

Thanks to Selector, DiBurning and UP2CSX

The repeated jerking/surging still takes place every few inches for only the middle half of the speed range. Doesn't happen when throttle is kept low and ends once the unit reaches 3/4 speed.

CV3 and CV4 inertia - Crandell suggests - these set the level of acceleration/ deceleration or inertia. Works in that unit takes longer to reach a changed speed setting. Seems to be no change in frequency or harshness of the repeated jerking/surging when in the mid speed range.

CV6 mid speed voltage - Crandell suggests - QSI doesnt seem to have this CV. Not shown in DecoderPro or manual.

CV29 28/128 speed steps - Jim suggests - no visible - change to 2(same as 34 with extended addressing) the opposite setting is 0( same as 32 with extended addressing). Not clear to me which is 28 (on or of) But no effect on the problem toggled either way.

Now I am wondering about decoder pro and the connection. A lot of messages come up at the bottom of the programming page indicating loco not found etc when trying to read ALL Sheets" but it is obvious some of it works bidiriectionaly because values are both retrieved and sent.

thanks ,,, ken

Next step is to figure out taking/uploading a video so you can see exactly the problem.
 
and The answer IS

BEMF or Back EMF Settings

I could not believe that this problem was mechanical since 2 areas of performance were so good.

I could not accept that the unit would have to go back for decoder repairs when apparently all functions worked - just some speed setting was causing haywire results.

I kept poking and thanks to suggestions and discusion here, I kept poking some more.

Nothing worked but I had always wondered about something called BackElectromotiveForce (I think). I believe it designed to keep speed performance consistent for a range no matter what load the unit is under.

Got more familiar with DecoderPro and Locobuffer and risked looking at the working F7 A unit. DP has a nice page for BEMF with 9 slider bars. Well all the speed related ones (8) were close to Zero for the A but on the B a slider called Medium Speed Differential Gain was close to 50%. Zeroed it and things are now running good. I imagine a lot of other things need to be checked for consistency but this is a good start. I guess the good question is 50% as a reset value? either reset still has problems or this is a very funny setting!

LocoBuffer and DecoderPro are everthing I hoped - could be improved but what an awsome tool. It would be another month if I had to solve this without them.

Medium Speed Differential Gain

thanks for everything ,,, ken
 
Last edited by a moderator:
and the QSI manual says

5.8.7 CV 56.18-21.SI: Quantum PID Parameters

“PID” is an acronym standing for “Proportional, Integral, Differential”. PID parameters are used to control the amount of power applied to the motor of QSI locomotives when operating under Regulated Throttle Control, Calibrated Speed Control, or Load Compensated BEMF Speed Control.

The PID control loop strives to match the locomotive’s measured BEMF to a target BEMF value. Basically, if the measured BEMF is less then the target BEMF, the PID control loop increases the power supplied to the motor to
make the locomotive go faster. If the measured BEMF is greater then the target BEMF, the PID control loop decreases the power supplied to the motor to make the locomotive go slower.

Because model locomotives behave differently at different speeds, we use four different sets of PID parameters within four different speed ranges. Since BEMF is directly proportional to speed, we classify these speed ranges in terms of BEMF ranges.

For Q1a firmware and Q2 firmware before 1-Oct-08, there are two PID parameters that you can set for each BEMF range. We do not allow changing the Integral Gain. For Q2 firmware after 1-Oct-08, you can also set the Integral Gain.

A relatively easy-to-read on-line summary of PID control can be found at the following site:
http://en.wikipedia.org/wiki/PID_controller


5.8.7.3 CV 56.20.SI PID Parameters for Medium Speed (PI=20, SI = 0,2)
Default Values: Depends on Locomotive
CV 56.20.0: Proportional Gain for: 54 < BEMF ≤ 108
Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0
D7 D6 D5 D4 D3 D2 D1 D0
• Use this CV to specify the Proportional Gain for medium speeds: 54 < BEMF ≤ 108.
• Valid values are 0 to 255.
CV 56.20.1: Integral Gain for 54 < BEMF ≤ 108
Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0
D7 D6 D5 D4 D3 D2 D1 D0
• Use this CV to specify the Integral Gain for medium speeds: 54 < BEMF ≤ 108.
• Valid values are 0 to 2. The decoder firmware may allow you to set any value 0 to 255, but during operation, if
the value is greater than 2, a value of 2 is used.
CV 56.20.2: Differential Gain for: 54 < BEMF ≤ 108
Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0
D7 D6 D5 D4 D3 D2 D1 D0
• Use this CV to specify the Differential Gain for medium speeds for: 54 < BEMF ≤ 108.
• Valid values are 0 to 255.

=======================================

Now the formatting is not nice like in the manual but compared to 8 sliders with labels it is still a big HUH????????
 
You must log in or register to reply here.



Back
Top