Boiler Water Analysis

There are two problems. One is that the RO is not performing. The other is that the dosing treatment regime is not performing. It needs a specialist in each of these areas to sort it out. Review the contractual arrangements in place on each part of the process and arrange for each specialist to pay a site visit; do this by telephone immediately.

I see many problems here, but would need more information about the steam generating system to provide a detailed analysis. For starters, the feedwater chemistry is terrible. My first suspicion is many impurities are entering the boiler feedwater from condensate return streams. And then if the chemistry chart is correct, the boiler water pH, among other issues, is terribly high. I have seen such conditions cause under-deposit caustic gouging, which can lead to tube failures in short order. Immediate action is necessary, but a more thorough analysis is not possible without additional system details. These include, as I mentioned, condensate return system arrangement and chemistry details, high-purity water system details and operational performance, etc. I recently wrote two articles for Industrial Water Treatment and Ultrapure Water magazines on performing plant water audits.

Brad: Since the OP is talking about a ~950 psig boiler in a sugar mill, there are a number of issues:

(1) Many sugar plants do not actually have a really high rate of condensate return, so they must rely upon a fairly substantial make-up rate. This could be anywhere from 320 gpm (if 0% return), to probably as low as 35 gpm demand.

(2) One of the problems where there is condensate return is contamination from the process, including such things as glycolic acid, sucrose, etc., and nearly all of the contaminants would produce an acidic reaction in the boiler, leading to low pH (not the present case), high TDS, and exceedingly high iron levels.

The current situation is this: The RO system is not producing a high enough quality of water for the system requirement, simultaneous with low rates of condensate return.

This results in having to turn up the condensate treatment (leads to high conductivity, even though the target pH is not reached), and this is further compounding the problem by inducing slight increases in boiler carry-over into the steam. There should be no phosphate in the boiler feed-water unless the chemical injection is taking place upstream of the sample point, and all chemicals are being added at that point.

It also appears that one boiler is being blown down more heavily than the other (difference in PO4 and conductivity). Both boiler pH values are excessive for the pressure range they are in. Iron is also a bit high in each, not good. I think they need to straighten out not just the boiler feed-water parameter range, but also the boiler brine chemistry with their vendor(s), and/or their local chemist if they have one.

Your insistence upon a water audit, is spot-on, because if the water budget is unknown, it makes diagnostics, and the solution a bit more tricky to arrive at.

Thanks for the info, James. I understand the situation better now. And you are absolutely correct, there is much chemistry that needs to be corrected, and soon.

1) What is the water condition at the RO discharge?

If you have not done so you need to determine if the RO unit is working correctly.

2) Why is the total dissolved solids (TDS) level in the water at 254?

From your table TDS and Silica values it appears there is an issue with the sand filter on U1 and/or the RO unit.

4) Is the RO and sand filter being operated at the correct GPM/LPM flow rate?

If the flow is not slow enough the RO unit cannot clean the water efficiently.

Too much flow through a sand filter will cause cavitation and will cause the silica to be entrained in the water stream.

3) How often do you blow down the boiler to clear collected sediment(s) and keep the TDS at a low value?

This should be occurring often enough to eliminate or at least minimize total dissolved solids in the water.

4) Do you have a competent, experienced water technician and/or engineer on board at the site to monitor the water condition?

If not I suggest you consider doing so because judging by the Fe (Iron) content in the water, the boiler and whatever equipment is being serviced by the boiler water is suffering from corrosion and will soon fail.

If you do not want to experience severe issues with the process and all of the associated equipment, you must obtain the services of a competent water treatment consulting/engineering company and follow their guidance faithfully.

(Bold for critical importance emphasis-Not shouting.)

A2) Sand filters only reduce TSS, and not TDS. Further, there should be nil TSS in RO permeate!

A4) The boiler blowdown point is a function of the pressure and the feed water treatment. The operation manual will surely give this detail.

Yes I agree that the sand filter is for TSS control and the RO unit is for TDS control.

However many water treatment facilities utilize sand filter(s) upstream of the RO to maximize the RO membrane life expectancy by removing as much TSS as possible before it enters the RO unit.

The high silica (non-dissolvable) readings are indicative that the site is using one in this instance. (I suppose their water source could be so bad as to cause the high readings but if so, the RO unit is either bypassed or the membrane has failed.)

I would be very surprised if the site does not have filter bypass and/or a RO bypass valves open.

Often site equipment is in such poor repair state that it suffers from high water leakage loss which in turn usually overtaxes the RO unit so they do partial loop flow treatment when making up water instead of treating all of the water that enters the system.

Because they will not shut the system down long enough to do the maintenance right and will not allow the RO unit enough time to clean the water properly it does not take very long before the water is fouled to the point it needs to be overboard dumped and new water installed.

I am guessing that this is the case.

There is no suggestion in the Original Post of a TSS problem.

There is no indication of the incoming water conditions, therefore any expression of upstream-of-RO treatment units is mere speculation.

The Original Poster is still well advised to review the contractual arrangements and performance criteria of the pre-treatment dosing and the original RO unit. Attempting this remotely over the internet with a bunch of anonymous strangers is abstruse in the extreme, particularly in the absence of Process criteria and commissioning records in that quarter.

An assessment of condensate quality would be a most valuable addition to the Original Poster's suppliers' information.

TDS ~ 0.55 Conductivity for mostly ionic solutes, although it can vary from one water to another slightly, 0.5-0.6 is a good approximate range.

The Silica being referred to is reactive soluble silica in the water, not sand particles. Silica can be soluble up to approximately 100-200 mg/L depending on the other conditions of the water.

A boiler of this pressure should have slightly better quality than what I expect they are producing, and probably needs a full demineralizing system in addition to the RO.

Boilers that are not being treated with suitable metal passivating agent, or on a controlled oxygen feed will under parasitic rates of corrosion, and depending on the holding time of water within the boiler brine (remember that boiling leaves the solids behind, unless there are level control issues, or cyclone separator issues), will have iron levels consistent with this, but may or may not have aggravated problems with internal corrosion that are sufficient to cause eminent tube failure. In fact, sometimes the iron level seen may just be a function of poor sampling technique, and included the presence of suspended corrosion product (magnetite, etc.). If this is a total iron test of unfiltered water, then it is a meaningless number. If the iron test is a membrane filtered sample, and still that high, I begin to scratch my head, since the pH is neither high enough or low enough to really support such high levels of soluble iron (unless contamination is causing the iron to be complexed and stay in solution.

This requires a specialist that can determine the onsite needs of your existing water, it is a black art that very few are effective at....it is a constant struggle...You usually arrive at a compromise that will dictate the maintenance schedule required that you feel is reasonable....The amount of time and effort spent on water quality is usually cut short by the amount of money you are willing to spend past the point where common sense says, this is too much.....so look for somebody that has a proven track record, and ignore the salesmen....

http://www.pdhcenter.com/courses/m165/m165content.pdf

The RO has a hole in it, that's plain to see.

Whether you do or don't repair it, is up to your budget.

You didn't state the age of the boilers.....

As an interim measure I would suggest increasing the blow down frequency and duration. Diluting the boiler brine thus will buy you time (defer feed water treatment costs) but cost you fuel.

With RO treated make up water your feed water dosing requirements would be minimal.

Flush regularly and get that RO sorted.

BTW, do you use condensate recovery in your process or is the condensate dumped? What is the condition of your make up water header tank ?

What is the balance of water purified and used/lost in the boiler and steam? How much blow-down is existing on each boiler if any?

How much chemical of each kind is being added to these boilers on a daily basis? You are on a chemical treatment plan, are you not?

What ancillary equipment is associated with each boiler? Do they have deaerating feed-water heaters?

I also notice an issue with one of the feed-water requirements: It seems very abnormal to allow up to 10 mg/L phosphate in boiler feed-water. Also, you need to introduce a chloride ion exchange vessel (also known as dealkalizer) upstream of the RO, to help with removal of bicarbonate alkalinity residuals that are distorting the water chemistry to lower value in the feed-water, and higher than desired in the boiler.

How often have you all ever checked for contamination of the return steam from process? What were the results?