E-Bolt: Double Acting Ram Operation

I have been ruminating on this for a while so I'll throw it out there for analytical discussion. There is a lot of effort to upgrade the GTA E-Bolt board to allow higher rates of fire than the 13 bps stock setting.

There are some who claim that the stock single acting ram cannot go any faster anyway. There are several places on the web showing how to modify the ram to make it double acting along with the installation of a 4 way solenoid to power the ram in forward and aft movements. The mods are supposed to be necessary to obtain the ROF available with a replacement card like the DM4/5/C.

I have another idea. Everyone knows that certain mods to the sear of a stock 98C can produce rates of fire in excess of 25 bps. This tells me that the blow-back pushes the hammer back quickly enough to the cocked position to allow for this very high ROF. My idea is to harness that blow back gas in either of two ways to provide for the high-speed return of the ram.

First idea was to modify the hammer end on the ram shaft to incorporate an "O" ring just like the stock bolt. The solenoid controlled ram drives forward to strike the poppet valve pin in the back of the power valve. When the poppet valve opens gas flows out and makes two right angle turns to go forward to get to the paintball. But, some of the gas leaks around the valve pin and exerts pressure on the hammer to blow it back to the cocked  position. This works well with the stock bolt, so with the lesser weight of the ram head it should throw it back to the "cocked" position at a higher rate of speed than it can the stock bolt...allowing for very quick cycle rates and the chance to see if my new Cheetah board can make my Halo actually give me 35 bps.

The other method would be to either modify the ram to be a two way or obtain one of the BSI two way rams. On the side of many/most power tubes is the port that supplies gas to operate the Cyclone Hopper. I would take that connection and install a tube to connect that gas pressure to the front of the ram. Thus as soon as the marker fires gas would be flowing to push the ram back to the "home" or "cocked" position.

Should be as-fast-as-fast-can-be. Tell me my flaws in the thought process?

Edited by Bruce A. Frank - 24 October 2006 at 4:49pm

Monk wrote: Might want to add a buffer somewhere on or in the ram as to not tear it up. And modding the RT fitting to the ram I think would only give you about 18-22 cps. Even then, you may not even get a complete cycle.

Now think about this, first off, I could leave the return spring in the modified stock ram so that the gas return pulse would be assisting the spring. The spring by itself can cycle the ram fast enough for at least 13 bps. Also realize that the air pressure supplied to the ram from the solenoid is less than 100 psi. The blow-back gas coming from the hopper port is going to be at least 200 psi and higher in most markers.

I do agree that a buffer "O" ring is likely needed and would install one on the ram shaft between the head and the body of the ram.

Monk wrote: Yeah, but the RT uses blowback and it can only cycle fast enough for 18-22 cps. I retract my may not cycle statement, I confused myself. lol.

Eighteen to twenty two would be all that I need. I am not looking to fire full, or RT or even burst. I have never fired my marker at any setting other than semi-auto. But, when I was getting a lot of play time in and practicing walking the trigger, I could occasionally get well ahead of the 13 bps of the GTA board.

Once I get the new board installed I will make these mods to the ram just to see of what the marker is fully capable.

I would suspect that the speed limitation of the Cyclone is related to the mass of the parts being moved. Also, the pneumatic cylinder operating the Cyclone suffers the same limitations as the standard E-Bolt ram...its return rate/speed is controlled by a spring.

The gas pressure supplied out of that port has to be more that the blow back pressure leaking past the valve pin.  And again, blow back can cycle a standard marker in excess of 20 bps. Seems only logical that using the pressure more directly would cycle the ram equally quick or even faster.

Edited by Bruce A. Frank - 25 October 2006 at 7:09pm

I see. Hmmm... It will definatly need a QEV.


EDIT: If you put a QEV on both inputs to the cylinder, then you could really get that thing moving fast. I think you might have one of the greatest ebolt mods right there.


Edited by Monk - 25 October 2006 at 11:25pm

I believe that Tippmann stuck with springs in the E-Bolt because it was less expensive.

Let me go into my operation method. I learned early on that it was easy to get the dwell just a little too long which interfered with ball feed. With just a slightly longer dwell it was possible to trigger a forward cycle of the ram before it was fully retracted from the previous cycle. At best this was preventing the bolt from coming back far enough to allow the next ball to drop into the receiver so no ball was fired.

At worst the bolt would move just far enough rearward for the ball to start to drop, but the next cycle triggered and the ball was chopped half way through the feed port. I worked and worked to shorten the dwell to the optimum level. Once there I have not touched it since. All of my velocity adjustment is now done with the secondary regulator (Palmer Stabilizer). Velocity can also be adjusted with the LP regulator that supplies pressure to the ram, but is limited to a maximum of 100 psi. (the solenoid valve starts to leak when pressure gets much above 100 psi)

With my dwell set as short as it is I thing I will not have a problem. There is another piece of information. The ram has no seal on the front end, so pressure pushing the ram piston rearward is not going to run up against a pneumatic lock. If the piston can't be pushed rearward the gas will likely vent around the unsealed shaft at the front of the ram.

Another thing is that since the solenoid leaks at pressures above 100 psi if the blow back gas were timed such that it impinged against the piston while the solenoid  was still energized I think there would just be leak down.

Monk wrote: Wouldnt a QEV solve all the problems?

It seems that it might solve any venting problem.

karll wrote: What board are you going to put on your ebolt? Are you running the GTA ebolt kit right now?

As I mentioned in another subject line, I have just been given a brand new TAG DM4/5/c board. (along with a Cheetah board for my Halo B) That will be the heart of my E-Bolt, but it is going to be a few weeks before I have a chance to put it all together. Right now I am looking for the correct connectors to use to attach my wires.

     I'm using Molex Picoblade connectors for my rig. So far I have only been able to obtain the components that make up the connectors and they are a pain in the butt to try to assemble into a working harness. Im trying to find a preassembled connectors but am coming up short. What are you using?

karll wrote: Found some info on this Bruce. A guy on Tippmann Owners said he tried this and it didn't work. Not a lot of info in that thread, but figured I would let you know. Thread.

OK, the likely reason there wasn't enough gas coming out to blow the hammer back is because the LPK valve pin fits MUCH closer to the rear bulkhead hole than does the standard valve pin. 

As you said there wasn't much detail there. I can't imagine that with a good fitting "O" ring mod to the ram's hammer head that there would not be enough force generated to push it back all the way to your elbow. Particularly when the head , rod and piston's total weight is about half that of the stock hammer (the hammer head on the end of the ramrod is smaller diameter than that of the stock hammer. It would require making a replacement that is larger in diameter to make an "O" ring fit. I will make an aluminum hammer with a steel striking button where it impacts the valve pin. A few calculations indicate that this aluminum hammer will be no heavier than the original solid steel one that comes on the pneumatic ram, requiring no more pressure to operate.)

Monk wrote: I also think the guy might have had the problem because (I think) he wasnt using QEVs.

I don't disagree with that.