Hoffmann point motor on DCC

[Ravenser]

I'm not sure if this is still a DCC question , as I am thinking of a hardwired solution but here goes....

I am attempting to finish the installation of point motors on my layout, Blacklade

Because of the unusual shape of the boards, there are some complications. Perhaps the best way to see the issues is if I post a photo of the underside of the boards:

Here is a view of the track on top , at an early stage in construction some time ago:



The station throat is right across the narrowest part of the boards - just 5" wide - and the board joint goes through the middle of the lot

And this is what you get underneath. Additional components needing to be installed have been placed in their locations:

There's an awful lot of point motors and wiring jammed into a narrow space. I had installed 6 motors, and baulked at the last 3 because thery were right in the neck and I literally couldn't get 3 tortoises to fit in. Since then I've sourced 2 more compact motors - a new Cobalt Blue motor bought at Ally Pally, and a Hoffmann motor. These can be seen on the top board, placed for a dry run to locate them , in the photo. The Hoffmann is right at the end - it will get the drive into a very awkward hole right in the side frame of the board, and avoids fouling the bolts linking the boards . Being long and narrow , it gets round the problems with Tortoise footprints

The Cobalt Blue is effectively a modern small Tortoise , and poses no issues. It runs off the same type of decoders in the same way.

The Hoffmann motor is rather different, and creates some issues. It is a 3 wire AC point motor, slow action motor drive. Consequently it can't be driven by the 2 wire 12 V DC outputs used by Tortoises. I tried connecting it to a traditional 12V DC controller to test- nothing happened (First query - I presume I haven't damaged it at all by doing this???) . See Finney & Smith's site: Hoffmann motor

I have the following accessory decoders:

- an NCE Switch It , currently fitted to worth the two Tortoises already installed

- a Digitrax DS64, requiring a 12V DC independant supply (someone reported on here I think that a 16 V AC external supply will in fact kill it despite the instructions) . This can be set to operate stall motors, or solenoids, with built in CDU. The advertising boasts "It's never met a switch machine it doesn't like!" Well, I think I've found one - Hoffmanns

- an unbuilt MERG decoder kit for steady state (ie motor driven point motors). This provides 12V DC output , 2 wire

None of these, as far as I can see , will drive a Hoffmann point motor

Other candidates are:

- NCE Snap It . This has a built in CDU , which I suspect will do a Hoffmann no good at all
- Digitrax DS52. This will control either 2 x stall motor point motors , or 2 x solenoid motors. If set to the latter, a built in CDU promises "a powerful capacitive discharge capability". Which again will I suspect fry a Hoffmann
- NCE and Digitrax decoders for Kato Unitrack. These are 2 wire, so almost certainly useless

- LS150 . The only decoder I know will drive a Hoffmann. But it's not that good for stall motors because the current shuts off after max 2 secs , and it costs £45. I don't need an expensive 6 output decoder - I intend to use the DS64 to work the 4 stall motor units, and so all I want is a cheap solution for one Hoffmann

After staring at the ceiling in dispair, I had a sudden inspiration - a piece of lateral thinking. The Hoffmann powers one half of what is really a crossover . The other half is driven by a Tortoise already installed. Good practice is to wire crossovers as a pair and drive them as one , eliminating conflicting movements , and using one decoder. As the two motors on this crossover are dissimilar types , you can't do this....

Or can you?

There are planned to be 3 main sets of circuits on the layout:

- The DCC traction buses (black+red)
- The auxilary circuits. There is a heavy blue/yellow circuit running down the layout from a Lenz/Roco 45VA 15VAC transformer. This powers the MERG decoder on one board, and I am installing a 16V AC to 12 V DC stablised supply converter on each board, powered from the heavy blue/yellow. This will feed 12 V DC to the Erkon signals, the Digitrax DS64 for points , and any lighting kits
- The circuit for 3 Kadee electro magnets

The idea is to tap 16V AC for the Hoffmann directly off the auxilary 16V AC supply circuit, and not use a decoder . Of course it then needs manual switching , and I would lose the ability to set up complete routes automatically using the Route macros on the Power Cab.

Except.....

One end of the crossover is powered by a Tortoise with two sets of switch contacts. One set of switch contacts is used to change the polarity of the frog - the other set is free, and not requirted for signalling.

If I connect the 16V AC feed to the input side of this set of switch contacts, and then connect the left and right feed contacts on the Hoffmann motor at the other end of the crossover to the output switch contacts on the Tortoise , operation of the Tortoise (through the DCC decoder) will switch the AC supply and operate the Hoffmann motor. Thus ensuring the 2 motors are electrically interlocked . The common contact on the Hoffmann is wired directly to one side of the 16V AC bus

The Tortoise contacts are rated to swich acurrent up to 1 amp, AC or DC . I take it that the current flow is determined by the draw of the motor, not the potential max output of the transformer?

Am I missing something , or have I cracked it, at a cost of a couple of short bits of wire???

[Crosland]

The Hoffmann motor is rather different, and creates some issues. It is a 3 wire AC point motor, slow action motor drive. Consequently it can't be driven by the 2 wire 12 V DC outputs used by Tortoises. I tried connecting it to a traditional 12V DC controller to test- nothing happened

I'm absolutely certain the motor in the hoffman is a normal DC motor (the price is a give-away). One side is common, one side is connected to the anode of a diode and the cathode of another diode. The other ends of the two diodes are the the two wires to the switch. The motor is actually running on half-wave rectified DC with the direction controlled by which half of the AC is rectified, depending which way the switch is connected.

 

It will run as is on DC, if you connect the two control wires together and connect (with the original common wire) to a DPDT switch, wired to reverse the DC supply to the motor. You do need some caution. Use a DC voltage the same or lower than the RMS voltage of the AC supply. The motor will be constantly driven, rather than being driven every half cycle so check that it isn't overheating and use the lowest DC voltage that works.

 

Andrew Crosland

[Ravenser]

I believe you are right about the actual motor being DC from something I've seen elsewhere

 

So to make this run on DC, one side of the DC supply is connected to the common, and the other side of the DC supply is connected to both the other ("L" and "R") terminals ? And then a means of reversing polarity is required?

 

My options are then

 

- run from the outputs of the same DCC decoder as the Tortoise (or from the NCE Switch It which would otherwise become redundant), with one side of the decoder output linked to the common on the Hoffman, and the other side feeding both the L and R terminals. The decoder switches the polarity of the DC supply at 12V

 

- hardwire for 15V AC off the auxiliary bus, with feed side switched through the spare contacts on the Tortoise at the other end of the crossover.

 

I'm still leaning towards the 15V AC option but it's helpful to know that I do have other options to drive directly off a decoder

[Guest baldrick25]

Using a Hoffmann point motor on a DC supply see page 4 of this PDF ( right at the bottom)

 

Hoffmann link

[eldavo]

The Hoffman link looks awfully complicated! I'm pretty sure you connect the two wires with diodes together then connect to this and the common connection exactly as if it is a Tortoise i.e. 2 wires with the DC feed being reversed by a switch or the point decoder. Certainly works with a Conrad motor which is the same mechanism.

 

Cheers

Dave

[Ravenser]

Thanks for all the help - but I'm a bit confused on one issue :

 

The Finney and Smith diagram shows a pair of diodes in the circuit. These are presumably external to the motor . The German instruction sheet with the motor doesn't - it shows the current wired directly from a switch to L and R. I think I've heard there may be a mistake somewhere

 

There's a practical concern here. Since I changed my job I no longer walk past a branch of Maplins in the lunch hour. Where I am now, the nearest electronic shop is probably 15 miles drive, and I don't have any diodes to hand

 

There is however one possible fudge to get some... The NCE Switch It instructions show 2 LEDs wired in opposite directions between decoder and Tortoise as an indicator. I have an Erkon ground signal, which was meant to be wired to the second set of contacts on the Tortoise to indicate the lie of the points . Could I use this as the necessary diodes in the Finney & Smith diagram - ie the SPDT switch is the spare set of contacts on the Tortoise and the 2 LEDs in the Erkon signal kit are wired in opposite directions between the 2 wires from the switch on the Tortoise?? NCE suggest aTortoise will act as a sufficient current limiter to protect the LEDs without resistors - the fact Finney and Smith

show a bipolar LED in the same indicator role under DC suggests the same applies to the Hoffmann

 

Or will the cycles of AC trash the LEDs?

 

It will run as is on DC, if you connect the two control wires together and connect (with the original common wire) to a DPDT switch, wired to reverse the DC supply to the motor. You do need some caution. Use a DC voltage the same or lower than the RMS voltage of the AC supply. The motor will be constantly driven, rather than being driven every half cycle so check that it isn't overheating and use the lowest DC voltage that works.

 

Andrew Crosland

 

So - if I connect to an NCE Switch It , with the common wired to one output of the decoder and the L and R terminals wired to the other, the decoder will change the polarity and drive the motor? Finney and Smith appear to show 2 transformers wired in a pair, and I don't have 2 transformers

 

Can you explain "RMS voltage"? I'm a bit of an electrical/electronic ignoramus... Would 12 V be too much??

 

 

 

[melmerby]

 

Can you explain "RMS voltage"? I'm a bit of an electrical/electronic ignoramus... Would 12 V be too much??

 

When using AC supplies as you may realise the voltage (and hence the curent) is constantly swinging positive and negative (Hence Alternating Current).

Root Mean Square is a way of calculating the AC equivalent (voltage/current/power) of a steady state i.e DC supply.

Simply it means that 12V AC RMS should pass the same equivalent current (and hence cause to dissipate the same power) in a component such as a resistance as would 12V DC. (It is not quite as straightforward as this but that's the general rule)

When using AC you have to remember that the peak is both plus and minus approx 1.4 times the quoted RMS value and any component (e.g. something like a non-polarised capacitor or a diode) being used on AC should have 2.8 times the voltage rating it would need on DC.

I hope that is not too confusing!

There is an article on Wikipedia but it's not that easy a read.

 

And 12V DC on a Hoffmann point motor (unless someone knows differently) should be fine.

As Andrew says try and find the lowest voltage it will reliably run on.

 

Keith

[jimhagan]

The Finney and Smith diagram is incorrect diodes are not required for a DCC points decoder which has AC outputs which is probably all DCC decoders intended for use with solenoid motors such as Hornby, ESU, Lenz, Traintronics, etc,etc.

 

People get confused about Conrad and Hoffman point motors believing they are the same design, apart from looking very similar the crucial difference is that Conrad motors have a DC motor inside so have diodes as required for an AC decoder used with a DC motor these are pre-fitted and must not be removed.

 

Hoffman motors have AC motors inside so do not required diodes to be fitted, but you have to provide your own wires so do not fit diodes, which is where the Finney and Smith diagram is wrong, in fact if you look at the Hoffman diagram there are no diodes fitted, but the diagram does not show a decoder, however it does show an AC input via a switch.

 

Both makes of point motor simply wire straight into a points decoder, common in the middle and left and right to either side of common, and in both cases the other three wires are for connecting to a live frog, in the case of the Conrad motor it is the yellow and two blue wires, use a meter if you are not sure which blue goes to which side of the frog.

 

I bought the Hoffman motors from Finney and Smith at an exhibition some years ago, and that time nobody on the stand could tell me if they could be connected to a decoder because they had never been asked the question!    

[Crosland]

I think there is some confusion here.

 

DCC accessory decoders output DC, either a pulse to drive a twin coil machine, a longer pulse for slow motion switch machines that do not have end of travel limit switches or steady state.

 

There seem to be two different types of Hoffman motor. The original was basically the same as the Conrad and uses both use a DC motor. I'm not at all  convinced that the later Hoffman motors have an AC motor internally. if that were the case, they would be completely incompatible with DCC accessory decoders. Much the same way Marklin's AC loco motors require special loco decoders.

 

The confusion with AC comes from the olden days when train set controllers would have an AC auxilliary output for operating accessories.

 

Andrew

[Ravenser]

As this has resurfaced from the depths, I suppose I should report that I ended up hardwiring a 15V AC supply off the auxillary bus, switched through the spare contacts on the Tortoise point motor , as originally intended. After 3 years' intermittent use (Blacklade doesn't get run as often as it should be...) there have been no noticeable problems.

 

The only issue to arise has been with the motor itself - at a certain point it "stuck" and buzzed rather than cutting off when thrown in one direction. I remounted it - it was mounted with double sided sticky pad , an idea pinched from the Tortoise instructions , though I wouldn't do it with Tortoises - and by ensuring it was slightly squarer the problem seems to have been resolved.

 

I know that on a club project I used to be involved with, Hoffmann motors were used with Lenz LS150s, and as far as I know with complete success