I have another recent thread on this issue, but I want to come at it from another angle.
My complaint is my three Intermountain HO scale F3A locos run too slow at the maximum speed setting.
All three locos have Digitrax DZ143PS decoders. At the time, the DZ143PS was the recommended decoder for the F3A by Intermountain.
I wrote to Intermountain about this speed issue and, and in their reply, they said that no one at Intermountain could recall the gear ratio for the F3. I thought that was strange. Their suggestion was to remove the decoder and do a speed test "in a DC environment". Easier said than done. They went on to say that normal performance can be achieved between 14 and 15 volts for DC or DCC. My DCC output is 14.7 volts. So, that response from Intermountain was not helpful.
On the previous thread that I had started, it was suggested that I turn off BEMF. I reset the values of CV54 (switching speed) and CV57 (BEMF) to zero, so both were turned OFF. Still, no difference in maximum speed.
So here is my question. Instead of the DZ143PS, would the other recommended Digitrax decoders for this loco, DN163PS or DH163PS, provide greater maximum speeds? Or would there be no difference?
Rich
Alton Junction
Rich,
I have a question for You. That output of 14.7, do you know for a fact, that the motor, is running on 14.7, or is the decoder, giving the motor less than that? Can you turn the throttle all the way up and measure, how many volts the motor is actually getting? If it is infact running on the whole 14.7, already, how many more volts will it stand before it burns up? Back to the gear ratio issue again.
Frank
Edit: Something to take a look at, if you want to take the time;
http://www.teamdavinci.com/understanding_gear_reduction.htm
That 14.7 volts is the reading of my DCC output voltage to the track, as measured by a RRampMeter, a circuit board that reads DCC voltage output.
I do not have a voltage reading to the loco motor or a reading under load.
I only mentioned it because of the reply that I received from Intermountain. I really don't care what voltage is reaching the motor. All of my locos have HO scale or N scale decoders installed and run a lot faster at maximum speed than these three Intermountain locos which are equipped with Z scale decoders.
My question is whether an N scale decoder will cause a loco to run faster than a Z scale decoder and whether an HO scale decoder will cause a loco to run faster than an N scale decoder.
Quote:
. I really don't care what voltage is reaching the motor. All of my locos have HO scale or N scale decoders installed and run a lot faster at maximum speed than these three Intermountain locos which are equipped with Z scale decoders.
I don't understand your statement. If the others are getting more voltage to the motor, would that not make them run faster, than the one that is getting less voltage? Why do you have N-scale and Z-scale decoders in a HO-scale engine, for size?
It's not getting less voltage to the motor because it is a "Z scale" decoder. The Intermountain motors are seeing the same peak voltage that any other loco is seeing, unless you've set a speed table or set CV5. The difference is either the motor that Intermountain uses, or the gearing, or both. A scale designation on a decoder is little more than a physical size reference, and sometimes a current capacity indicator, though many "Z scale" decoders have the same current ratings as a similar "HO scale" decoder. The big large scale decoders, those often have a 4 amp or higher rating. The motor drives of most any decoder I've ever heard of are all the same, transistor H bridge switching rectified track voltage to the motor, so the peak voltage that the motor gets is really set by the DCC system's track voltage. Peak can be reduced by setting a speed curve or adjusting CV5, but that's it. BEMF can possibly 'reserve' a bit at the top end to allow it to increase motor power under load, even if already at full throttle. Switching speed is extremely obvious - when engaged, default uses F6 on Digitrax decoders, it runs the motor at half speed, and also reduces any momentum effects in CV3 and 4 by half.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
Randy, thanks, that's what I was looking for.
It may be, though, that another brand of Z scale decoder may provide a higher speed at maximum setting than the Digitrax Z scale decoder. True?
zstripe Quote: . I really don't care what voltage is reaching the motor. All of my locos have HO scale or N scale decoders installed and run a lot faster at maximum speed than these three Intermountain locos which are equipped with Z scale decoders. I don't understand your statement. If the others are getting more voltage to the motor, would that not make them run faster, than the one that is getting less voltage? Why do you have N-scale and Z-scale decoders in a HO-scale engine, for size? Frank
Frank, yes, the Z scale decoder was recommended by Intermountain for space reasons, although a comparable N scale decoder would fit as well. An HO scale decoder would be a tighter fit.
What I meant by my reference to voltage is that I only measured it since Intermountain raised it as a potential issue. But, as Randy confirmed, the voltage reaching that particular motor should be similar regardless of whether the decoder was Z, N, or HO.
Re-read Randy's reply.
If the DZ143PS is putting full track voltage (minus H-bridge rectification losses) to the motor, how will using a different brand of decoder be able to put more than full track voltage (minus H-bridge rectification losses) to the motor?
Sure, there may be some minor difference in those rectification losses due to different components used by the different manufacturers, but I don't think it would be nearly as much as what you've described with your F3's vs. your other locos.
As an example, I have 18 Atlas GP's of various vintages that have Digitrax decoders in them: DH123PS's, DH163A0's, a DH163IP, DN135PS's DZ123PS's, and DZ125PS's - Whatever was cheap and/or on-hand at the time.
Now we KNOW that all those different form factors use different H-bridge components, yet all those locos run close enough to the same speed that I haven't bothered to speed match any of them.
So I agree with Randy that it's probably either the loco's gearing, or you have a CV setting that's limiting top speed.
Stevert Re-read Randy's reply. If the DZ143PS is putting full track voltage (minus H-bridge rectification losses) to the motor, how will using a different brand of decoder be able to put more than full track voltage (minus H-bridge rectification losses) to the motor?
Stevert, I was not saying that using a different brand of decoder would deliver more than full track voltage to the motor. My understanding of Randy's comments is that any size decoder, Z, N, or HO, will deliver the same voltage to the motor and no more than the track voltage.
What I then asked was whether another brand of Z decoder might provide a higher speed at maximum setting than the Digitrax Z scale decoder. It has been suggested that Digitrax decoders may run slower due to their CV configurations and values. For that reason, I am asking whether a different brand of Z scale decoder would result in the Intermountain loco running faster than with the Digitrax Z scale decoder installed.
My last comment.. If both engines, the fast and the slow, are getting the exact same voltage to the Motor, the slower one has a different gear ratio, than the faster one. What else is left?
I think that it was Randy in my other thread who mentioned that at least some Digitrax decoders run slower than other brands with comparable decoders. If I recall correctly, it is because of the way that Digitrax uses CVs. i am hoping that Randy will have something more to say in this regard.
zstripe My last comment.. If both engines, the fast and the slow, are getting the exact same voltage to the Motor, the slower one has a different gear ratio, than the faster one. What else is left? Frank
No, nothing different about Digitrax - most decoders that have BEMF will reserve a little bit of the top end if BEMF is turned on - otherwise is the load increased while at full throttle, it wouldn;t be able to increase motor power at all. However, BEMF at high throttle is more of a crusie control than anything, that's why I like TCS, since the BEMF turns off above a certain speed step. But turning off BEMF would eliminate that.
Also, with the previous Digitrax sound decoders, the microcontroller just isn't powerful enough to play multiple sounds AND calculate BEMF, so when you hit thehorn, they would slow down - unless you turned off BEMF. Totally different issue.
richhotrain Stevert Re-read Randy's reply. If the DZ143PS is putting full track voltage (minus H-bridge rectification losses) to the motor, how will using a different brand of decoder be able to put more than full track voltage (minus H-bridge rectification losses) to the motor? Stevert, I was not saying that using a different brand of decoder would deliver more than full track voltage to the motor. My understanding of Randy's comments is that any size decoder, Z, N, or HO, will deliver the same voltage to the motor and no more than the track voltage. What I then asked was whether another brand of Z decoder might provide a higher speed at maximum setting than the Digitrax Z scale decoder. It has been suggested that Digitrax decoders may run slower due to their CV configurations and values. For that reason, I am asking whether a different brand of Z scale decoder would result in the Intermountain loco running faster than with the Digitrax Z scale decoder installed. Rich
You're missing the point.
At it's highest speed step, the Digitrax decoder delivers full track voltage minus rectification losses, to the motor
At it's highest speed step, a different brand will deliver full track voltage minus rectification losses, to the motor.
How will that make the loco run faster?
Randy, I have a question, even though it does not really seem like one at all, and would like your opinion on it.
Back when I was first getting into DCC, about 1999, we had a gentleman in the club who was a beta tester for Digitrax. We had been putting some DZ12x's, can't remember if 121's or 123's into some HO loco's, mostly P2K geeps, and we noticed that after a period of time, 6-10 months, the decoders started acting up. They would then eventually fail.
We asked our Digitrax buddy why this was happening, and this is what he said. That while the Z-scale decoders had the same rating as HO decoders, the Z-scale decoders heat sink couldn't get rid of the heat generated by the electronics fast enough due to its size as opposed to heat sinks in larger decoders. The Z-scale decoders were literally cooking themselves into failure. I know that even Soundtraxx recommends that you attach an extra heat sink to their TSU-750 decoders due to this.
We replaced the Z-scale decoders with DH123's and they have been fine since "upgraded".
What do you think?
Carey
Keep it between the Rails
Alabama Central Homepage
Nara member #128
NMRA &SER Life member
Stevert At it's highest speed step, the Digitrax decoder delivers full track voltage minus rectification losses, to the motor At it's highest speed step, a different brand will deliver full track voltage minus rectification losses, to the motor. How will that make the loco run faster?
I went back and looked at the replies to my previous threads on this issue. In one reply, a remark was made that full voltage may not be getting to the motor due to lighting, circuit board, decoder settings, CV values, etc.
Stevert, I am not saying that a different brand of decoder will make the loco run faster. I don't know. Since I am no expert on this issue, that is why I am asking the question.
All I can say is that my observations and conducting speed trials around the layout tell me that these three Intermountain locos are running considerably slower than any other loco on the layout. To me, it is either the decoder or the gear ratio on the loco. What else could it be?
richhotrainAll I can say is that my observations and conducting speed trials around the layout tell me that these three Intermountain locos are running considerably slower than any other loco on the layout. To me, it is either the decoder or the gear ratio on the loco. What else could it be?
i'm no expert either, but not all 12V motors run at the same max speed at max voltage. There may be a trade off between speed and torque. See the 12V RPM column on the NWSL motors catalog page.
greg - Philadelphia & Reading / Reading
cjcrescent Randy, I have a question, even though it does not really seem like one at all, and would like your opinion on it. Back when I was first getting into DCC, about 1999, we had a gentleman in the club who was a beta tester for Digitrax. We had been putting some DZ12x's, can't remember if 121's or 123's into some HO loco's, mostly P2K geeps, and we noticed that after a period of time, 6-10 months, the decoders started acting up. They would then eventually fail. We asked our Digitrax buddy why this was happening, and this is what he said. That while the Z-scale decoders had the same rating as HO decoders, the Z-scale decoders heat sink couldn't get rid of the heat generated by the electronics fast enough due to its size as opposed to heat sinks in larger decoders. The Z-scale decoders were literally cooking themselves into failure. I know that even Soundtraxx recommends that you attach an extra heat sink to their TSU-750 decoders due to this. We replaced the Z-scale decoders with DH123's and they have been fine since "upgraded". What do you think?
It's possible, since they are amsller. But if you weren't exceeding the rating, they shouldn;t fail. The only Digitrax decoders with an actual heat sink are the heavy duty large scale ones, none of the others have a heat sink of any sort under the shrink wrap. In the early days, there were soem runs of P2K E units and/or PA's that had the wrong motor installed and drew somewhere north of 3 amps, which would fry even HO decoders. Never heard of the Geeps getting that motor though. The TSU-750 is just a bad design, rated at less than 1 amp and it still overheats unless plenty of airflow is provided ot the thing is held against a large heat dissipating surface. The Loksound Micros are significantly smaller and have a higher current rating, and do not overheat. Part of it has to do with the way the main chip is made - like modern CPUs being made on a smaller process node, like 22nm instead of the older 45nm, a modern chip can run faster then the older ones yet generates less heat. I suspect this is main issue behind the TSU-750 heat problems, remember the Tsunami is an old design, downright ancient in terms of how fast complex chip design progresses. That they have heat issues despite including a heat sink in the package is testament to an inefficient design.
As for the older Z and N scale decoders - if you're talking 121 and 123 series, those are some of the first of the smaller decoders, and even though were rather large compared to the current line of Z scale decoders. I think they had a lower current rating than the modern ones, designs have gotten better. So have the features. Again, there should never be a problem unless you exceed the rating, which is why they have one. They usually have 2 ratings, one is the continuous, the other is the max it can handle on a temporary basis - the loco needs to be tested to make sure the motor does not exceed the continous rating. I rarely use the smaller decoders - I use the largest one that will fit, mainly because they are cheaper. When there's room for a DH or DN, not sure why I would want to spend more to put a DZ in there - actually I only use TCS, but the same thing 0 I use the T1 whenever possible, only a few of my locos are so limited in space that I need to use the MC series to save space. Let alone the even smaller M series (no connector on the decoder - the MC's have a smaller version of the 9 pin connector found on the T1).
richhotrain Stevert At it's highest speed step, the Digitrax decoder delivers full track voltage minus rectification losses, to the motor At it's highest speed step, a different brand will deliver full track voltage minus rectification losses, to the motor. How will that make the loco run faster? I went back and looked at the replies to my previous threads on this issue. In one reply, a remark was made that full voltage may not be getting to the motor due to lighting, circuit board, decoder settings, CV values, etc. Stevert, I am not saying that a different brand of decoder will make the loco run faster. I don't know. Since I am no expert on this issue, that is why I am asking the question. All I can say is that my observations and conducting speed trials around the layout tell me that these three Intermountain locos are running considerably slower than any other loco on the layout. To me, it is either the decoder or the gear ratio on the loco. What else could it be? Rich
So have you checked the voltage going to the motor at the highest speed step?
Have you checked to see if the board has resistors, etc. which might restrict that voltage?
Have you done a CV8=08 reset, which also resets the speed tables, to eliminate a CV setting?
Have you turned off BEMF (CV57=0), which is known to have an effect in some cases (especially when in a consist, which would seem to apply here)?
Have you checked the gear ratio? I know you got an inconclusive reply from Intermountain, but it's easy enough to check it yourself. Just use a sharpie to make a mark on a flywheel and on a drive wheel (the tread is a good place that won't show later). Then turn the flywheel by hand and count how may revolutions it takes for the drive wheel to complete a full revolution.
Or have you convinced yourself that a different brand of decoder is the only cure, and you just want someone to confirm that for you?
Stevert So have you checked the voltage going to the motor at the highest speed step? Have you checked to see if the board has resistors, etc. which might restrict that voltage? Have you done a CV8=08 reset, which also resets the speed tables, to eliminate a CV setting? Have you turned off BEMF (CV57=0), which is known to have an effect in some cases (especially when in a consist, which would seem to apply here)? Have you checked the gear ratio? I know you got an inconclusive reply from Intermountain, but it's easy enough to check it yourself. Just use a sharpie to make a mark on a flywheel and on a drive wheel (the tread is a good place that won't show later). Then turn the flywheel by hand and count how may revolutions it takes for the drive wheel to complete a full revolution. Or have you convinced yourself that a different brand of decoder is the only cure, and you just want someone to confirm that for you?
No, no, yes, yes, no, and sorry that you have become such a wiseguy. You seem determined to ridicule my basic question on this thread. Reply after reply by you has mis-stated what I have asked since my initial post.
Is this decoder connected via an 8 pin plug? If so, just wawap one of your others. Don't care if it doesn't physcially fit, just run it with the shell off. That will clear up any doubts about the decoder. I still say it is simply that hey have geared them down more than the others, though ti could be a different RPM motor as well, or both. No amount of decoder tinking will change that.
To reiterate, there is nothing about the Digitrax DZ series decoders that would make them put less voltage to the motor compared to a similar series DN or DH decoder with the same settings. And they do all take the same settings across the line, ie, all Dx125 use the exact same CVs and values, the only difference is physcial size and number of functions.
rrinker Is this decoder connected via an 8 pin plug? If so, just wawap one of your others. Don't care if it doesn't physcially fit, just run it with the shell off. That will clear up any doubts about the decoder. I still say it is simply that hey have geared them down more than the others, though ti could be a different RPM motor as well, or both. No amount of decoder tinking will change that.
Thanks, Randy, I will do that, swap out the Z decoder for an 8-pin HO decoder that I have on hand. Good idea.
And, while I have the shell off, I will count the teeth on the gears to establish the gear ratio.
Thanks for your answer Randy. These were club locos, and all had been tested as continous draw and stall current as well. I never understood why you'd put in a z scale decoder into an engine where a cheaper HO decoder would work. I personally prefer Digitrax, DH 123's as 99% of my locos are steam and I don't need anything else other than 2 functions and I can get them cheaper than TCS, as I'm on a limited income.
As I said, I was just learning DCC at the time, and we were aware of the higher amp draw motors on the E's and PA. We had replaced those motors with some from NWSL, and never lost a decoder in them. The GP's lost every z-scale decoder eventually installed. I replaced them as they failed with DH123's.
I just never understood the logic they were using at the time, and still don't.
cjcrescent I never understood why you'd put in a z scale decoder into an engine where a cheaper HO decoder would work.
I never understood why you'd put in a z scale decoder into an engine where a cheaper HO decoder would work.
My only defense is that I was following Intermountain's decoder recommendation at the time.
And there's really nothing wrong with that. I doubt those Intermountain units draw more than a 1/2 amp with the motor stalled, well within the limits of the DZ123 or DZ125. Intermountain isn't the only one who suggests Z or N scale decoders for their HO locos - it's really just a size differentiation, not much different electrically.
Not sure on why the lack of durability on the Proto Geeps. I have 4 of them, don't think they draw over 1 amp stalled, which is the continuous current rating for the DZ series decoders. But mine are all of approximately the same vintage, with the upgraded drives over the initial P2K releases, and are from the cracked gear era, which is why I just repalced all of them with Athearn gears BEFORE the stock ones cracked. I am not aware of the ridiculous high current motor problem extending past the E's and PA's, though the motor is the same physical size and appearance and could easily be swapped to a Geep chassis.