Accumulator testing w/pressure gauge

[2seater]

I mentioned in another thread that I was going to try to get actual data on the performance of the accumulators I have on the shelf. My test rig utilizes my pressure gauge assembly for testing system pressure on the car, a 500ml graduated cylinder and a used pump/motor assembly. The graduated cylinder is just a cheap plastic Amazon special accurate to +- 5ml with a hose barb tapped into the bottom as the reservoir. I used the same 350ml level as the start of each test to minimize variation. I ran the pump manually with a battery on the bench and stopped at test pressures I chose based on what I believe are close to correct: 1500psi for red dash light turn off, 2000 psi for pump restart via pressure switch and 2600 psi for pump stop. Accumulators A & B are used from my '89 and '90 and C is a new stock one I installed briefly and then replaced with the Spinning wheels larger model. I won't go into the reasons for replacement but there was a clear difference. I ran each test three times but they were very consistent:

 

A: 170ml @ 1500psi, 190 @ 2000 and 200 @ 2600   also 208ml @ 3050 which is the ball pressure rating

B: 155ml @ 1500psi, 175 @ 2000 and 192 @ 2600           202ml @ 3050

C: 60ml  @ 1500psi, 102 @ 2000 and 135 @ 2600            150ml @ 3050

 

As can be seen none of them held close to the .250ml @ 210 bar on the rating stamped into the ball. I believe that is the total volume of the ball from rough measurements of the exterior. When testing, the pressure immediately jumped up to close to 1000 psi upon pump start for ball C, which seems to indicate a full nitrogen pre-charge, much longer for the other two. Please note, all of these accumulators were operational and not causing real issues when removed. One more thing, both the pressure switch and ball were just run down by hand without mechanical assistance as the sealing is done by an o-ring, so no need for gorilla force.

 

The one takeaway I found was the amount of fluid contained between 2000 psi pump on and 2600 psi is pretty small but significantly greater in the fresh accumulator:

A=10ml, B=17ml and C=33ml

 

I don't have my cars out of storage so I cannot define what the precise fluid level change would be in the reservoir, using the amount of fluid listed above, but less drop is definitely better based on what I found.  I should add, as long as there is significant run time, less is better, but a fully flooded one will show very little change in level and a short run time. If there are any suggestions on where I could improve or other types of tests, I welcome the comments.

 

Edited April 8, 2017 by 2seater (see edit history)

[Barney Eaton]

Being old, I can relate to oz better than ml, (ml will give more accurate readings, as few graduated containers are in oz)

Looking at "C" the volume from 0 (depressurized) to 3050 psi (150ml = 5.07 oz)

The worst you had was "A" and it was a total of 7.03 oz

If this was measured in level drop in the reservoir (which I will simulate with water on a spare I have) I hope this will give an idea of visual comparison in the reservoir.

If you have a known bad accumulator, it's numbers would be interesting

[2seater]

1 hour ago, 89RedDarkGrey said:

Could you please briefly explain the "bottom line" conclusions-

 

1) Why are used accumulators holding more fluid than a NOS unit? Less precharge= room for more fluid?

Yes, as far as I can tell, that would be true.

1 hour ago, 89RedDarkGrey said:

 

2) Why does the larger SW unit hold the LEAST amount of fluid? MORE precharge= less room for fluid?

I may have been misleading in the way I worded this. The C unit is an almost new standard accumulator. It was installed on my '90 and when I found little change in the number of "pumps" required to start the pump, I ordered the larger SW accumulator. I have never been able to pass the accepted brake tests and yet I know my system is working at the FSM working pressures? Perhaps it is what qualifies as a "pump" is my stumbling block? That is what I meant by that being another story.

 

3) Which model IYO is "better" and why?

I imagine this is related to #2. The C unit is clearly better in the fluid capacity within the normal working range. It may be true the older accumulators may offer a little longer partial boost, but I have no way to quantify how rapidly the boost would deplete as the pressure drops. The SW unit is on my car which is still in storage and I will test that when I return it to service. I have no data on that at this time.

 

It would seem that a larger amount (volume) of pressurized fluid available in the circuit- would be better?

Yes, I agree. I noted the brakes feel more firm with the SW unit. I suspect the pressure depletes less quickly with greater volume so the "feel" is better, but no data to back it up.

1 hour ago, 89RedDarkGrey said:

 

 

Maybe install an AMP clamp (meter) on the pump, and run until amperage goes up- would indicate the moment of bladder force against the nitrogen precharge. That could make a type of reading/calculation.

I will see about that. I estimated the pre-charge by the initial pressure spike when the pump turns on. It flows quite a bit of fluid in open flow and I always flowed fluid through the system with the drain valve open to be sure there was a minimum of trapped air when changing the accumulator, before I ran the actual test with the valve closed. I did think about attempting to check the pump run times but I figured it would be gibberish as the supply voltage in the field would likely be different plus as a one man operation, and running the pump manually, accurate timing would be difficult

 

Thank you for your work on this topic

 

[2seater]

2 hours ago, Barney Eaton said:

Being old, I can relate to oz better than ml, (ml will give more accurate readings, as few graduated containers are in oz)

* Yes, I know what you mean. The graduated cylinder is in milliliters, and I almost used cc's in my post, which would also be generally correct. The accumulator ball capacity is marked as 0.25l, or 250ml, I figured using the same units would reduce errors. Our engines are one gallon or four quart capacity if you want to confuse people

Looking at "C" the volume from 0 (depressurized) to 3050 psi (150ml = 5.07 oz)

The worst you had was "A" and it was a total of 7.03 oz

* Yes, that looks correct, however, the maximum working pressure is approx. 2600 psi so the working fluid drop would need to be recalculated. I only tested to 3050 psi as that is the maximum listed on the ball next to the volume capacity to see how close it came to the volume rating.

If this was measured in level drop in the reservoir (which I will simulate with water on a spare I have) I hope this will give an idea of visual comparison in the reservoir.

* I would have done the fluid level change if my car was available, but that was not to be. Any additional help in that regard would be great.

If you have a known bad accumulator, it's numbers would be interesting.

* I do not have a totally dead one and I checked all I have on hand. By the way, I tried Ronnie's probe test, and could find no discernable difference in the location of the diaphragm even on a well used one.

 

[Barney Eaton]

I would be using the reservoir.....under no pressure when I use water or any other cheap fluid to convert the drop in the reservoir to ounces or milliliters.

Simply stated..if you measure 3/8 inch drop in the reservoir level, what does that convert to in ounces?

Then you would compare that with Hal's findings with the accumulators.

[Ronnie]

Seems like an easy way to measure the amount of fluid stored in the accumulator - with it on the car - would be to:

1. Install the accumulator ball you want to test
2. Make sure the fluid level in the reservoir is exactly on the full mark (or a better mark you want to make yourself)
3. Turn on the pump and let it run until it stops.

Then fill the reservoir back up to the full mark using a container calibrated in CCs or MLs like Hal described earlier. That would allow you to keep track of the amount of fluid you have to add to bring the fluid back up to the full mark. Reading the amount of fluid removed from the calibrated container would tell you how much fluid is stored in the accumulator ball on the car with it under normal operating conditions.

[2seater]

I think I see what Barney has in mind and is similar to my thinking. What I was trying to do is get a better handle on the actual volume of fluid that charges the accumulator with the aim of transferring that to a reservoir reference level change. I know we have an approx. standard developed from years of actual experience from forum members and my purpose was to quantify. Part of the issue is the actual shape of the interior of the reservoir with the level gauge is unknown to me and the measuring point needs to be consistent for good confidence in the results. The brake fluid doesn't seem to have much of a meniscus in the measuring cylinder so I would think the surface tension is lower. Water might work fine but the surface tension would need to be knocked down with a couple drops of detergent. I guess I would try alcohol for level testing. If on the car, obviously brake fluid is the liquid of choice.

[2seater]

The vacuum is a good idea, everyone has one of some sort. I made a vacuum unit from an old two piston automotive a/c compressor and a castoff one gallon bug sprayer jug as my fluid reservoir/catch can. A good shop vac. should be able to pull maybe 8"Hg of vacuum, so that should work just dandy. If run for long, do you bleed in a little external air to keep the vacuum motor cool. Short term shouldn't be an issue. Would work okay to pull fluid through the front brake bleeders too. I know my forearm gets mighty tight pumping that MityVac hand pump when doing that.

[DAVES89]

2seater I just picked up those two pump/motors and we could test those accumulators if you want. Give me a call or email if you are interested.

[Barney Eaton]

I tried measuring the fluid level change in the reservoir today.  I used rubbing alcohol for the fluid...used that for a couple of reasons

(1) there is a filter at the output and if I used water, I would get water in the brake fluid if I ever used this reservoir.

(2) alcohol is less messy and even a little dirty it can be used to flush brake lines (3) it is cheaper than brake fluid.

I have photos but even with backlighting the fluid level does not show up well in a photo. I taped a scale to the side of the reservoir and filled it to a reference point on the scale (started at 1/2 inch on the scale)

Initially each ounce of fluid = 1/16 inch on the scale, when I got to 5 oz the level had risen 7/16 inch so each oz is slightly more than 1/16 inch.

Since Hal has not tested a known bad accumulator we will need to wait for that and see if the number I got relate.

From my experience, a 1/2 inch drop in reservoir level is borderline bad......meaning you probably will not have problems stopping under normal conditions but a panic stop may give a hard pedal and a red brake light.

Edited April 11, 2017 by Barney Eaton (see edit history)

[2seater]

I believe you are on the right track, and that is my ultimate goal; turn the fluid volume measurement into something everyone can use as a reference. The present standard is good. My intention is to tweak it a bit. The figures I get for the three accumulators I tested convert to 4.56, 6.59 and 6.76 oz. from empty to fully charged at pump cutoff of 2600psi. I listed the capacity at rated pressure of the ball @ 3050psi also but in normal use it never charges to that pressure. This was done just for reference but perhaps is misleading? All of the ones I tested were still operational and the brakes felt normal, however I couldn't pass the brake pump test before the pump turned on. Even the new ball was good for no more than two "pumps". If memory serves, the used accumulators were in the 9/16" drop range and the new one was a bit under 1/2". I wasn't being very critical at that time so the measurements should be taken with a dram of ethanol

[2seater]

      I am bringing this thread back as I finally got my car home to complete this run of testing. I moved the first three accumulator observations to this post and added three others. Accumulator "D" is from my beater '89,  "E" is Ron's Hydac which is almost new on loan to me in transit to its new home and then "F" is my two year old Hydac. I added the number of "pumps" required for pump restart and the observed fluid drop for each accumulator.

Fluid capacity at various pressures                    Estimated pre-charge  Fluid drop in reservoir   Pumps to pump restart   Working pressure fluid

 

A: 170ml @ 1500psi, 190 @ 2000 and 200 @ 2600      425#                              .698"                                   1 pump                              10ml 

B: 155ml @ 1500psi, 175 @ 2000 and 192 @ 2600      500#                              .670"                                   1 pump                               17ml

C: 60ml  @ 1500psi, 102 @ 2000 and 135 @ 2600       1050#                            .475"                                  2+pump                             33ml          

D: 170ml @ 1500psi, 190 @ 2000 and 200 @ 2600       425#                              .698"                                    1 pump                              10ml

E: 107ml @ 1500psi, 155 @ 2000 and 192 @ 2600        950#                              N/A                                        N/A                                  37ml

F: 107ml @ 1500psi, 155 @ 2000 and 193 @ 2600        925#                              .682"                                    3 pump                              38ml

 

The observed fluid drop closely mimics the accumulator capacities noted in the pressure testing. While the Hydac units are larger, the two I have tested appear to have a somewhat lower gas pre-charge than a new standard style resulting in greater useful working fluid capacity, but not as much as indicated by the size. I have never been able to pass the "pump" test, so next on the agenda is a fluid change and all new flex brake hoses to see if that makes any difference.

Edited May 31, 2017 by 2seater (see edit history)

[2seater]

I replaced the front brake hoses which were cracked and flaking the outer casing just above the intermediate clamp. I decided not to do the rear flex lines. They are not subject to the twisting the fronts are and upon experimenting a bit, it appears the steel lines may twist off, even though they are rust free. Not in the mood to remake them at this point.

 

I did get the flushing completed this morning and there does seem to be a marginal improvement in the number of pumps before pump restart. I found it now takes (4) pumps consistently. The other item is it takes the full (25) pumps to get a hard pedal when pumping the accumulator down. With the standard accumulator I generally got 15-18 but the larger Hydac has consistently shown a greater reserve, maybe not at normal braking levels, but an increased safety factor.

By the way, the car has 137k miles on it.

[DAVES89]

2seater came over today as I had 4 each Teves pump/motors with unknown quality accumulator balls and pressures switches. His testing setup is something to see. I could see that going forward with his testing he would need the connectors that come with these units and as luck would have it I had all the connectors he would need going forward.

 So back to the project at hand. I had three pump/motors that could be used on the later 1988-1990 Reatas and one early 1988 pump/motor. The later one we could not test, which was fine as I knew the motor ran and we were able to switch out the accumulator ball and pressure switch and test them. So out of all out testing [which was quick and easy] I ended up with 4 good pump/motors, 4 good pressure switches and 1 good accumulator ball. I did give the early 1988 pump/motor to 2seater so he has an extra motor for future use as compensation for his efforts. The pump due to a different hose used in the early 1988 Reattas is no good to him [or me].

 I will leave him to post the results that he found and how it ties into the information he listed with his previous tests.

[2seater]

Yes, Dave and I ran through several pump accumulator combinations today. This was my first try with the three indicator lights connected, yellow abs, red pressure and blue for pump switch, which turned out to work pretty well. Thanks to Dave I now have the plugs to connect pump and pressure switch wiring more easily and quickly. I won't detail the results specifically but I did notice the trends of estimated gas pre-charge remain consistent with the amount of fluid pumped in. While we did not find a totally depleted accumulator, there was one that was down in the 150-200 psi range and held the most fluid @ 217ml. and had <12ml within the working range. It definitely would trip the pump on with approx., one pump or less. Two were in the sort of barely acceptable range of 400-500#. We did find one with over 800# of charge and would be acceptable as a decent backup while a new one was procured. This one should yield 2+ pumps and contained about 25ml. in the working range.

 

In conjunction we also tested four pressure switches and including the three tested before, on and off car, all seven total turn the pump on and off within the book values of 2000psi and 2600psi. I have not found a failed pressure switch yet. The other parts of the switch operate the yellow and red warning lights. These were slightly better than  the beater switch I was  using on my test pump, but fall just slightly low on the pressure required to turn the lights on and right at the low value to turn the lights off. In my opinion they were all good operational switches with the pump operating pressures spot on and the indicator lights very slightly out of range. A small caveat would be that observing the light and pressure at the same time is very difficult as there is always a lag between seeing the light and checking the associated pressure. With this almost doubled sample size, the original conclusions are confirmed and predictable.

 

As Dave can attest, the pumps run a long time before any real action happens on the pressure gauge with a failed accumulator, so it would seem the wear on the pump isn't just from the lack of reserve but also from excessively long run times on startup.

[Barney Eaton]

Glad to see you guys are deep into understanding the pressure/volume comparison and have a way to statically test the Teves parts.  

I am getting more calls from people that cannot find replacement parts, so if you can find used parts, test them, you can help keep Reattas on the road.