A tip on wiring the rheostat:

If you're going to solder the connections (which is a good idea), strip the wire going to the wiper (the center lug) extra long, and tie it to the otherwise unused side lug. Solder the wire to both lugs. The other wire goes to the third rheostat lug, as usual.

This makes absolutely no difference in normal operation; but if you are adjusting the rheostat with power applied, and the wiper bounces a little, the resistance only jumps to 50 ohms (the full rheostat value) instead of to infinity when the wiper is in the air. In addition to reducing noise in the circuit, which is pretty irrelevant here, it can reduce sparking between the wiper and rheostat, which prolongs its life.

mini wrote:.......
5. I agree that a true pre-infusion isn't possible. At the highest resistance, flow is still fairly rapid, and I don't think it's a good idea to add more.
...........................
Keep up the discussion, it has been really beneficial to me

The reason you still have a high flow even at highest resistance is pressure; The pump will still deliver almost full volume at low pressure. (below 2-3bar) It will, however, stall or quit pumping at a much lower pressure.
This low pressure/high volume behavior makes it tough to do a pre-infusion.

When you start the pump for a shot, the pressure is low for the first few seconds.
As soon as the pressure ramps (after any internal volume in the brew path is filled with water and can no longer be compressed, full brew pressure is developed across the puck.)

Having a boiler/brew pressure gauge on the machine will really help you to understand what you are dealing with in your tests.


Edit: I just ran some quick tests with a variac on Venus; it may not be so bad for pre-infusion pressures...

With 70volts to the pump, it stalls at 4 bar, while at 2 bar it delivers about 1.5 oz in 30 seconds.
...So start your shot at 70volts for the first 4-5 seconds, then quickly ramp it up to 108 for the duration of the shot.

For Gaggias (or at least my older Classic), the pump doesn't run when the steam switch is engaged, so you can't use the steam switch to turn on the heater during a shot.

I may just stick in a few batteries to supply to voltage to turn on the SSR.

Randy

Lots of posts! And some very good ones.

ira wrote:You really need to adjust the frequency if you want to solve the pre-infusion problem.

I concur. To get very slow pre-infusion flow rates out of a vibratory pump (with a pure electrical mod*) you would need to start decreasing the oscillation frequency. However, besides the natural limitation of effective frequencies, Ulka pumps run strictly at 60Hz using simple half-wave rectification of line current. Re-arranging the wave is beyond my current expertise and free time, and chopping it at hard angles can cause huge resistances in the solenoid. I believe rheostats are safer for your pump than simple lamp dimmers which reduce overall power in this manner.

kschendel wrote:A tip on wiring the rheostat:

A good tip! I had read about doing that before, but forgot. I'll probably run a jumper across the two terminals pretty soon.

Frost wrote:So start your shot at 70volts for the first 4-5 seconds, then quickly ramp it up to 108 for the duration of the shot.

mini wrote:I plan to test a lot of shots starting at this lower pressure end and then increasing to my desired "set point." I see no reason not to make the initial flow as gentle as possible

I win?

I realize that distribution, dose, tamping, etc. are all the usual culprits in channeling. I also realize that if I practiced enough I could get good, solid pucks at almost any pressure. But, as discussed in the how-to, I really think that this mod will make my machine a lot less critical of puck prep. Why waste coffee while I am still learning?

There are too many factors for me to test currently. I also received a digital scale just before I did my mod, so experimenting with temperature isn't in the fore front of factors currently.

I'm still in the process of getting my bearings with the mod, but everything has been very positive so far. By starting the shot at low pressure, and then quickly ramping up to slightly under full pressure, I have been getting some really nice shots. Blonding has been delayed much longer than my average on all four of the shots done with this technique. The shot I pulled tonight seemed to flow extremely evenly and blond at an almost perfect time. channeling seems to be mitigated a bit as well.

Tomorrow I will do another comparison test - full pressure vs. ramp up - in a little more detail. If anyone has some quick tests they would like to hear about, let me know. This weekend I plan to churn out some coffee, as well as make some videos. Until then, results will come rather slowly I'm afraid.

The column of I's is pretty crazy, isn't it! That was completely unintentional.

Consistency is tough to get down, and more so with these machines. Figure out how to get your machine to consistently deliver 9bar 200F shot after shot. Then get your grind and dose tuned. Then when you can pull 3 shots in a row and all turn out the same..... I'm getting there.

Getting the pump tuned from 11+ bar to around 9 bar makes a big difference for my shot quality. (after much experiment I settled on 8.5-9 bar range for my taste) I will likely order one of those rheostats to have on hand, and maybe install some day, but I have not been so motivated to implement variable pressure for pre-infusion. On the Venus, the pressure ramp starting a shot is quite gentle.

I'm wondering if it might be useful for cafe crema type extraction at much lower pressures too....
For coffees that espresso fails to deliver.

The shot with full pressure tasted good tonight. I've been using higher doses and seeing improvement. I was using Toscano - managed to get some nice caramels, and the slight bitterness was likely because I let it run too long.

The shot with soft, then almost full pressure tasted a little bit better, but not very much. It was more chocolate, with a pretty good creaminess. It was less bitter, but two shots is hardly a good test. The only scientific difference I noted was in the crema consistency. For whatever reason, the second shot had denser, smoother crema. Perhaps it has less gas because of the lower pressure? The mouthfeel was nicer than the first, for sure.

I pulled both shots with exactly the same prep - dose, distribution, tamp, boiler timing, etc. The second flowed more evenly, but both probably had some pinhole channeling.

I'm still enjoying the mod. It is easy to play around with mid-shot and opens up a lot of possibilities for experimentation. One thing to re-affirm is that the pump ramps up slower at higher resistances - at least according to my ears. Whereas normally maximum pressure is reached in about 3 seconds on my machine, at the highest resistance setting the pump takes ~6 seconds to reach full pressure.

A few observations: My pump (a 1991 Ulka) still functions at 75 ohms of resistance, which produces a pretty low pressure. However, it's hard to get an exact pressure reading with the blank disk in because the pressure continues to slowly rise at any resistance level. Is this normal?

I'll post the results of some tasting later today

Does your machine have an OPV? I thought that the pump pressure would stabalize pretty quickly at either max pressure or OPV release. I guess if there is truly zero flow, then the pump might do something different...

______________________________

Well, it turns out that my snapshot camera is not adequate enough to film espresso shots in a very meaningful way. Especially since I don't have a botomless PF. So you will have to take the following comment as objective without proof, based upon my 5 month espresso machine ownership averaging a drink every day.

I'm getting better shots.

With great care and attention to detail, I pulled about 10 shots this weekend. I'm almost positive that 2 of them were among the best 5 shots I've ever pulled with my equipment. And all of them were pretty good. Granted, I'm weighing every dose now, and I'm trying to distributing a bit more evenly too. But I'm telling you, I really think this pressure mod is helping.

My machine is simply more forgiving than it used to be. The overall occurance of channeling has been reduced, and when I do see channeling it is less catastrophic. This, to me, makes it worthwhile. The voltage mod isn't some miracle cure, but the the benefits are ceratinly worth the price and effort required.

I'm interested to see if you feel the same way, Frank.

I don't have an OPV (but I've ordered one). That probably explains why the pressure doesnt level off. It does level off while pulling shots though. I pulled 7 doubles this weekend to test out the mod. However, I didn't get the grind dialed in until the last 2 or 3. Therefore, it's hard for me to make any firm conclusions yet. However, the last double that I pulled was one of the best I've pulled. I have noticed that the crema appears to be lighter in color and denser at lower pressures. I think channeling has been reduced too, but I don't have a bottomless portafilter to know for sure. Unfortunately I only have access to my espresso gear on the weekends, so more testing will have to wait. I plan to do some side by side comparisons at full pressure and reduced pressure this weekend.